HIGH AFFINITY NUCLEIC ACID LIGANDS OF CYTOKINES
FIELD OF THE INVENTION
Described herein are methods for identifying and preparing high-affinity nucleic acid ligands to cytokines. The method utilized herein for identifying such nucleic acid ligands is called SELEX, an acronym for Systematic Evolution ofLigands by Exponential enrichment. This invention specifically includes methods for the identification ofhigh affinity nucleic acid ligands ofthe following cytokines: IFN-gamma, IL-4, IL-10, TNFα, and RANTES.
Further disclosed are RNA ligands to IFN-gamma, IL-4,IL-10, and TNFα. Also disclosed are DNA ligands to RANTES. Specific examples are provided of
oligonucleotides containing nucleotide derivatives chemically modified at the 2'-positions ofpyrimidines. The oligonucleotides ofthe present invention are useful as
pharmaceuticals or diagnostic agents.
BACKGROUND OF THE INVENTION
Cytokines are a diverse group ofsmall proteins that mediate cell
signaling/communication. They exert their biological functions through specific receptors expressed on the surface oftarget cells.
Cytokines can be subdivided into several groups, including the
immune/hematopoietins, interferons, tumor necrosis factor (TNF)-related molecules, and the chemokines. Representative immune/hematopoietins include erythropoietin (EPO), granulocyte/macrophage colony-stimulating factor (GM-CSF), granulocyte
colony-stimulating factor (G-CSF), leukemia inhibition factor (LIF), oncostatin-M (OSM), cilary neurotrophic factor (CNTF), growth hormone (GH), prolactin (PRL), interleukin
(IL)-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-9, IL-10, and IL-12. Representative interferons (IFN) include IFNα, IFNβ, and IFN-gamma. Representative TNF family members include TNF α, interferon (IFN) β, gp39 (CD40-L), CD27-L, CD30-L, and nerve growth factor (NGF). Representative chemokines include platelet factor (PF)4, platelet basic protein (PBP), groα, MIG, ENA-78, macrophage inflammatory protein (MIP)l α , MIP1 β, monocyte chemoattractant protein (MCP)-1, 1-309, HCl4, C10, Regulated on Activation, Normal T-cell Expressed, and Secreted (RANTES), and IL-8.
IFN-gamma
IFN-gammawas first described 30 years ago as an antiviral agent (Wheelock, 1965). Since thattimethe proteinhas been shownto be aremarkably pleiotropic cytokine whichplays importantroles inmodulating virtually all phases of.immune and
inflammatory responses. The cDNAs formurine IFN-gamma (Gray and Goeddel, 1983) and humanIFN-gamma (Gray and Goeddel, 1982) have beencloned, sequenced, and characterized.
IFN-gamma is amemberofafamily ofproteins related bytheirabilityto protect cells fromviral infection. This family has been divided into three distinct classes based on avariety ofcriteria, IFN-alpha (originally known as Type I IFN orLeukocyte IFN), IFN-beta (also originally known as Type I IFN orFibroblast IFN) and IFN-gamma (originallyknown as Type II IFN orImmune IFN). IFN-gamma is unrelatedto the Type I interferons at both the genetic and protein levels (Gray et al., 1982). The human and murine IFN-gammaproteins display a strict species specificity intheir abilityto bindto and activate humanandmurine cells. This is due at least inpartto theirmodest homologies atboththe cDNA and amino acid levels (60% and40% respectively).
IFN-gamma is produced by a unique set of stimuli and only by T lymphocytes and natural killer (NK) cells. The human andmurine genes forIFN-gammaare 6 kb in size, and each contain four exons and three introns. These genes have been localized to human chromosome 12 (12q24.1) andmurine chromosome 10. Activation ofthe human gene leads to the transcription ofa 1.2 kb mRNAthat encodes a 166 amino acid polypeptide (Derynck etal., 1982). The humanprotein contains a23 residue amino terminal hydrophobic signal sequence which getsproteolytically removed, giving riseto amature 143 residue positively charged polypeptide with apredicted molecularmass of 17 kDa. Variable post-translational enzymatic degradation ofthe positively charged carboxy terminus (Rinderknecht etal., 1984) isresponsible forthe charge heterogeneity ofthe fully mature molecule. Proteins with six different carboxytermini have been detected forboth natural and recombinant forms of IFN-gamma. Two polypeptides self-associate to form a homodimerwith an apparent molecularmass of34 kDa (Scahill etal., 1983). The homodimer is the biologically active form ofthe protein. Mature human IFN-gamma contains no cysteine residues, thus the homodimer is held together entirely by noncovalent
forces. This quaternary structure ofthe native protein explains its characteristic sensitivity to extremes ofheat (protein denatured at temperatures above 56°), and pH (activity rapidly lost at pH values less than 4.0 and greater than 9.0) (Mulkerrin and Wetzel, 1989).
The remarkable pleiotropic effects ofIFN-gamma are mediated through binding to a single type ofIFN-gamma receptor. The structure and function ofmurine and human
IFN-gamma receptors have been described (Schreiber et al., 1992). These receptor proteins are expressed on nearly all cells (except erythrocytes), and platelets (Anderson et al, 1982). The receptor binds ligand with high affinity (Kd = 10-9 - 10-10M) and is expressed on most cells at modest levels (200 - 25,000 sites/cell). Upon IFN-gamma . binding to the receptor at the cell surface, the intracellular domain ofthe receptor is phosphorylated at serine and threonine residues (Hershey et al., 1990).
One of-the major physiologic roles ofIFN-gamma is as a regulator ofimmune response induction, specifically its ability to regulate expression ofclass I and II major histocompatibility (MHC) antigens on a variety ofimmunologically important cell types (Trinchieri and Perussia, 1985) Functionally, IFN-gamma dependent upregulation of MHC gene expression is an important step in promoting antigen presentation during the inductive phase ofimmune responses and may play a role in antitumor activity of
IFN-gamma (Buchmeier and Schreiber, 1985).
Another major physiologic role for IFN-gamma is its ability to activate human macrophage cytotoxicity (Schreiber and Celada, 1985). Therefore, IFN-gamma is the primary cytokine responsible for inducing nonspecific cell-mediated mechanisms ofhost defense toward a variety ofintracellular and extracellular parasites and neoplastic cells (Bancroft et al., 1987). This activation is a result ofseveral distinct functions of
IFN-gamma. IFN-gamma has been shown to effect the differentiation ofimmature myeloid precursors into mature monocytes (Adams and Hamilton, 1984). IFN-gamma promotes antigen presentation in macrophages, through the induction ofMHC class II expression as described above, but also by increasing levels ofseveral intracellular enzymes important for antigen processing (Allen and Unanue, 1987). Macrophage cell surface proteins such as ICAM-1 are upregulated by IFN-gamma, thus enhancing the functional results ofthe macrophage-T cell interaction during antigen presentation
(Mantovani and Dejana, 1989). IFN-gamma activatesthe production ofmacrophage derived cytocidal compounds such as reactive oxygen- and reactive nitrogen-intermediates andtumornecrosis factor-a (TNF-a) (Ding etal., 1988). IFN-gamma also reduces the susceptibility ofmacrophagepopulationsto microbial infections.(Bancroftetal., 1989). Animal models have beenused to study the role ofIFN-gammainthe clearance of microbial pathogens. Neutralizing monoclonal antibodiesto IFN-gammawere injected into mice before infectingthemwith sublethal doses ofvarious microbial pathogens. These mice losttheir abilityto resolve the infection initiated withListeria monocytogenes (Buchmeierand Schreiber, 1985), Toxoplasmagondii(Suzuki etal., 1988), orLeishmania major (Green et al., 1990).
Besides these nonspecific cell mediated cytocidal activities, IFN-gamma also enhances othermacrophage immune response effectorfunctions. IFN-gamma
up-regulates expression of Fc receptors onmonocytes/macrophages (FcgRI), thus enhancingthe capacity of the macrophage for antibody dependentcellkilling (Erbe etal., 1990). IFN-gamma also promotes humoral immunitythrough enhancement of
complement activity. Itdoesthis intwo ways, i) bypromotingthe synthesis ofavariety ofcomplementproteins (ie., C2, C4, and FactorB) by macrophages and fibroblasts, and ii) byregulating the expression ofcomplement receptors onthe mononuclearphagocyte plasma membrane (Strunk et al., 1985).
IFN-gammaalso exerts its effects on other cells ofthe immune system. It regulates immunoglobulinisotype switching onB cells (Snapperand Paul, 1987).
IFN-gammaplays apositiverole inthe generation ofCD8+ cytolytic T cells (CTLs) (Landolfo et al., 1985) and enhances NK cell activity. Recently, ithas been unequivocally establishedthat CD4+ T cells do not constitute ahomogeneous class ofcells. Indeed, a paradigm oflymphokine biology and ofthe function ofCD4+ T cells has arisen, the so-called Thl/Th2paradigm (forareviewsee Paul and Seder, 1994). The THIclones, through theirproduction ofIFN-gamma, are well suited to induce enhanced microbicidal and antitumor activity in macrophages as detailed above (enhanced cellularimmunity), while the Th2 clones make products (IL-4, IL-5, IL-6, IL-10, IL-13) that are well adapted to act in helping B cells develop into antibody-producing cells (enhanced humoral immunity). The role played by IFN-gamma at this crucial nexus ofT cell effector function
is fundamental to the success or failure ofthe immune response.
IFN-gamma plays a major role in promoting inflammatory responses both directly, and indirectly through its ability to enhance TNF- α production. During an inflammatory response, cells leave the circulation and migrate to the point ofinfection. During this process they must first bind to and then extravasate through vascular endothelium. Both IFN-gamma and TNF-α can promote the expression ofoverlapping sets ofcell adhesion molecules (ICAM-1, E-selectin, and others) that play an important role in this process (Pober etal., 1986; Thornhill etal., 1991). In fact, experiments have shown that these two cytokines exhibit synergistic effects in up-regulating cell adhesion molecules in vivo (Munro etal., 1989). One can envision microbial infections in which the microorganism is already widespread at the time the immune response develops or in which the response does not quickly rid the host ofthe infectious agent. This results in continued T cell activation inducing both local inflammation and tissue damage with ensuing loss of normal function. Indeed, when the infectious agent is oflittle intrinsic pathogenicity, the disease induced by the infection may largely reflect the consequences ofsuch a response.
Excessive production ofIFN-gamma may play a role in autoimmune disorders (for review see Paul and Seder, 1994 and Steinman, 1993). The mechanism for this may involve excessive tissue damage due to aberrant or enhanced expression ofclass I and class II MHC molecules or the role ofan excessive TH1 cellular response. A role for IFN-gamma and the tissue-damaging effects ofimmune responses mediated by THl-like cells has been suggested in autoimmune disorders such as rheumatoid arthritis (Feldmann, 1989), juvenile diabetes (Rapoport etal., 1993), myasthenia gravis (Gu etal., 1995), severe inflammatory bowel disease (Kuhn etal., 1993), and multiple sclerosis (Traugott, 1988).
IL-4
Interleukin-4 (IL-4) is a remarkably pleiotropic cytokine first identified in 1982 as a B cell growth factor (BCGF) (Howard etal., 1982). In that same year, IL-4 was identified as an IgGl enhancing factor (Isakson etal., 1982). Because ofthe effect IL-4 has on B cells, it was first called BCGF-1, later termed BSF-1 (B-cell stimulatory factor-1), and in 1986 it was given the name IL-4. The cDNAs for murine IL-4 (Noma et
al., 1986; Lee etal., 1986) andhumanIL-4 (Yokota etal., 1986) have been cloned, sequenced, and characterized.
IL-4 can be regarded as theprototypic member ofafamily ofimmune
recognition-induced lymphokines designatedthe "IL-4 family" (forareview see Paul, 1991). This family consists ofIL-4, IL-5, IL-3, and granulocyte-macrophage
colony-stimulating factor (GM-CSF). Theproperties shared by theseproteins leads to this grouping and include, i) the linkage ofthe genes forthe members ofthe family (van Leeuwen etal., 1989), ii) the actionofeachmemberofthe family as ahematopoietic growthfactor in additionto any effects itmay exert on lymphoid cells, iii) the receptors forthese proteins are all members ofthehematopoietin family ofreceptors (Bazan,
1990a), and iv) coexpression ofthese factors bya subpopulation ofcloned CD4+ T cells (the so-called TH2 cells) (Mosmann etal., 1989) and bymastcells (Plautetal., 1989).
Theremarkablepleiotropic effects ofIL-4 aremediatedthroughbindingto cell surfacereceptors (IL-4R). ThemurineIL-4R(Mosely etal., 1989; Haradaetal., 1990), andthehumanIL-4R(Idzerdaetal., 1990; Galizzi etal., 1990) have beencloned, sequenced, and characterized. IL-4Rarepresentonavariety ofhematopoietic (Parketal., 1987) andnonhematopoieticcells (Lowenthal etal., 1988). Onbothhuman and murine resting T andB cells, IL-4Rarepresentinlownumbers (<400) andare regulatedby cytokines andotherfactors. ThereceptorbindsIL-4withhighaffinity (Kd = 10-10M). Nowthatmostofthereceptors forimmunoregulatory andhematopoietic cytokines have been cloned, itis apparentthatthemajority ofthese receptors fall into alarge family. This hematopoietic cytokine receptor superfamily includes receptors forIL-4, IL-2 ( b and g chains), IL-7, IL-9, andIL-13 whichmodulatethe lymphoid system; and receptors for erythropoietin, granulocyte-colony stimulating factor(G-CSF), GM-CSF, IL-3, and IL-5 whichmodulate thehemopoietic system. The superfamily also includes receptors for factors believed to normally function outsidethe immune and hematopoietic systems, including receptors forgrowth hormone (GH), prolactin, leukemia inhibitory factor (LIF), IL-6, IL-11, and ciliary neurotrophic factor (CNF) (forareview see Kishimoto etal., 1994).
A general first step in the signaling processes ofimmune and hematopoietic cytokines may be ligand-induced dimerization ofreceptorcomponents whose cytoplasmic
regions interact to initiate a downstream signaling cascade. The IL-4 receptor has a long putative intracellular domain (553 amino acids in mouse, 569 in human) with no known consensus sequences for kinase activity or for nucleotide-binding regions. The
biochemical nature ofsignals induced by the binding ofIL-4 to its receptor have not been elucidated. It does appear that the cytosolic domain ofthe receptor is essential for its signaling function (Mosely etal., 1989). Ligand induced dimerization ofthe IL-4 receptor appears to be a critical first step in IL-4 mediated signal transduction.
One ofthe major physiologic roles ofIL-4 is as a B lymphocyte activation and differentiation factor (Rabin etal., 1985; Oliver etal., 1985). The protein was first isolated based on this activity. In this regard, IL-4 activates production ofIgGl (Vitetta et al, 1985), but is also responsible for isotype switching in B cells from production ofIgG to IgE immunoglobulins (Coffinan etal., 1986; Lebman and Coffman, 1988, Del Prete et al, 1988). The effect ofIL-4 on the in vivo regulation ofIgE has been clearly
demonstrated. Neutralization ofIL-4 by treatment with a monoclonal anti-IL-4 antibody (Finkelman etal., 1986) or a monoclonal antibody to the IL-4 receptor (Finkelman etal., 1990) will block the IgE response. A recombinant soluble IL-4 receptor has been shown to inhibit IgE production by up to 85% in vivo (Sato etal., 1993). IL-4 deficient mice produced by gene-targeting in murine embryonic stem cells have normal B and T cell development, but serum levels ofIgGl and IgE are strongly reduced (Kuhn etal., 1991). IL-4 augmented IgE production leads to an atopic state (allergy/asthma) (Finkelman etal., 1989; Katona etal., 1991).
The IL-4 mediated up-regulation ofIgGl may play a role in the inflammation cascade. IgGl has recently been shown to form immune complexes which bind to the cellular receptors for the Fc domain ofimmunoglobulins (FcRs) leading to an
inflammatory response (Sylvestre and Ravetch, 1994; Ravetch, 1994). IL-4 transgenic mice have been produced that hyperexpress IL-4 (Tepper etal., 1990). These mice have elevated levels ofserum IgGl and IgE and they develop allergic inflammatory disease. These findings demonstrate the critical role IL-4 plays in the humoral immune response.
Another major physiologic role for IL-4 is as a T lymphocyte growth factor (Hu-Li etal., 1987; Spits etal., 1987). IL-4 enhances the proliferation ofprecursors ofcytotoxic T cells (CTLs) and their differentiation into active CD8+ CTLs (Widmer and Grabstein,
1987; Trenn, 1988). IL-4 appears to augmentthe IL-2 driveninduction of
lymphokine-activatedkiller (LAK) cells (Higuchi etal, 1989), whichhave been shownto lyse avariety oftumor cell targets without majorhistocompatibility complex (MHC) restriction. The roleplayedby IL-4 atthis crucial nexus ofT cell effectorfunction is fundamental to the success orfailure ofthe immune response.
IL-4 has been shownto affectnonlymphoid hematopoietic cells in avariety of ways. IL-4 has been shownto modulate monocyte/macrophage growth (Mclnnes and Rennick, 1988; Jansen etal, 1989) while enhancing theirdifferentiation and cytotoxic activity for certaintumor cells (Crawford,etal., 1987; Te Velde etal., 1988). IL-4 also has activity as a stimulant ofmastcell growth (Mosmannetal., 1986; Brown etal, 1987), and increases production andrecruitment ofeosinophils (Tepperetal, 1989).
IL-4 alone or in conjunctionwith othercytokines canpromote the expression ofa variety ofcell-surface molecules onvarious cell types with diverse implications for disease. Specifically, IL-4 can interactwithtumornecrosis factor (TNF) to selectively enhance vascular cell adhesionmolecule-1 (VCAM-1) expression contributing to T cell extravasationat sites ofinflammation (Briscoe etal, 1992). IL-4 alone orincombination withTNF orIFN-gammahas been shownto increase both MHC antigen and
tumor-associated antigen expressionon avariety ofneoplastic cells (Hoon etal., 1991).
As detailed above, IgGl immune complexes bindto the cellularreceptors forthe Fc domain ofimmunoglobulins (FcRs) leading to an inflammatory response. Inhibition of IL-4 andthe subsequentreduction inIL-4 mediated IgGl expression may prove efficacious against immune complex inflammatory disease states. Indeed, inhibitory ligands to IL-4 might also preventthe IL-4 mediated overexpression ofVCAM-1, thus attenuating the ability ofT cellsto extravasate at sites ofinflammation.
Inhibition ofIL-4 activity with amonoclonal antibody, arecombinant soluble IL-4 receptor, or gene knock-out, results in areduction ofserum IgE levels. An inhibitory oligonucleotide ligand to IL-4 could be clinically effective against allergy and allergic asthma.
A recentreporthas described adisorder in bone homeostasis in transgenic mice that inappropriately express IL-4 underthe direction ofthe lymphocyte-specific proximal promoter forthe lckgene (Lewis et al, 1993). Bone disease inthese mice resulted from
markedly decreased bone formationby osteoblasts, features identical to those found in humanosteoporosis. Inhibitingthis IL-4 mediated reductionin osteoblast activity may prove beneficial against osteoporosis.
Graft-versus-host disease (GVHD) is amajorcomplication ofhumantissue transplantation. GVHD does not exist as a single clinical manifestationbut can involve immunologic abnormalities ranging from immunodeficiency to systemic autoimmunities (Ferrara et al., 1991). These systemic autoimmunities include clinical and serological manifestations ofhuman systemic lupus erythematosus (SLE). Several murine models of SLEhave been developed (Gleichmann et al., 1982; vanRappard-van DerVeen et al., 1982), andthe induction ofsystemic GVHD inmice has been described (Via et al., 1988). Two recent studies have shown in vivo efficacy ofamouse monoclonal antibody to IL-4 in preventing GVHD and SLE inthese murine model systems (Umland et al., 1992;
Ushiyama et al., 1995). These observations suggestthataninhibitorofhumanIL-4 may be effective intreatment ofchronic systemic autoimmunities suchas SLE and GVHD.
A variety ofmicrobicidal infections are characterized by depressed cellularbut enhancedhumoral immune responses, which suggests aTH2 type ofresponse to infection. This TH2 phenotype is characterized by T cell secretionofIL-4, as detailed earlier. IL-4 blocks the microbicidal activity ofIFN-gammaactivated macrophages infighting
Leishmania major infection (Liew et al., 1989; Leal et al., 1993). Inhibition of IL-4 would enhancethe TH, effectorarm of theimmune response enhancing cellular immunity and leading to the resolution ofinfection. Neutralization of IL-4 in vivo allows mice otherwise susceptible to Leishmania major infectionto fight offthe parasite and clearthe infection (Heinzel et al., 1989). Several informative studies have looked atthe TH1ITH2 phenotypic distinctionin infectedmice, and suggestaTH1 dominatedresponsebeing most effective in fighting microbial infection (for areview, see Sher and Coffman, 1992).
IL-10
IL-10 is acytokineproducedbythe Th2 cells, but not Thl cells, and inhibits synthesis ofmost ofall cytokines producedby Thl cells butnot Th2 cells (Mosmann et al., 1991). Inadditionto the effect on CD4+ cells with Thl phenotype, IL-10 also inhibits CD8+ T cells with "Thl-like" phenotype. IL-10 is apotent suppressor ofmacrophage activation. Itcan suppress the productionofproinflammatory cytokines, including TNFα , IL-1, IL-6, IL-8 andIFN-gamma. Overall, these results suggestthatIL-10 is apotent macrophage deactivatorand an effective anti-inflammatory reagent. In addition, IL-10 prevents the IFN- g -induced synthesis ofnitric oxide, resultingin decreased resistance to intracellularparasites (Gazzinelli et al., 1992).
Bothhuman andmouse (hIL-10 and mIL-10, respectively) have been cloned and expressed (Moore et al., 1990; Vieira et al., 1991). Thetwo cDNAs exhibit high degree of nucleotide sequencehomology (>80%)throughoutand encode very similar openreading frames (73% amino acidhomology). Bothproteins are expressed as noncovalent homodimers that are acid labile (Moore et al., 1993). Whethermonomers are equally bioactive is not clearyet. Based onthe primary structure IL-10 has been categorized into the four a-helix bundle family ofcytokines (Shanafelt et al.,1991). Possibly due to high degree ofsequence homology and similar structure hIL-10 has been shownto be active on mouse cells (Moore et al., 1993) but notvice versa. hIL-10 is an 18 kDapolypeptide with no detectable carbohydrate; however, inmIL-10 there is one N-linked glycosylation. The recombinanthIL-10 has been expressed in CHO cells, COS7 cells, mouse myelomacells, the baculovirus expression system and E. coli. The rIL-10 expressed inthese systems have indistinguishable biological behavior (Moore et al.,1993).
Parasitic infectionoftenleadsto polarized immune response ofeitherThl orTh2 type which can mediate protection or susceptibility. The outcome ofaparasitic infection depends onthe nature of theparasite and the host. The best understood example is Leishmaniamajor infection in mice. L. major is aprotozoanparasite that establishes an intracellular infection in macrophages, where itis mainly localized in phagolysosomes. Activated macrophages can efficiently destroy the intracellularparasite and thus parasitic protection is achieved by macrophage activation. Nonactivated macrophages do not kill these organisms. As expected, activation ofmacrophages upon IFN-gamma treatment
enhanced the protection, whereas IL-4 and IL-10 blocked the increased microbicidal activity induced by IFN-gamma (Liew etal., 1989). In most inbred strains (example, C57/BL6) cutaneous infection ofL. major often leads to localized infection with spontaneous healing and confers resistance to reinfection. However, in BALB/c mice, L. major infection induces nonprotective immune response by producing IL-4. The antibody response mediated by IL-4 is ineffective and leads to death (Howard etal., 1980). In healing strains a strong Th1 response has been noticed with high level ofIFN- g, whereas in susceptible BALB/c mice a nonproductive Th2 response with significant levels ofIL-4 was found (Heinzel etal., 1991). Further it was shown that a single injection of monoclonal anti-IFN-gamma antibody can convert a resistance into a susceptible mouse (Belosevicetal., 1989). As expected, the treatment ofBALB/c mice with anti-IL-4 antibody led to the development ofTh1 response and healing (Sher & Coffrnan, 1992). Thus, depending on the nature ofthe pathogen, changing the immune response to a T cell subset with a protective phenotype can lead to therapeutic intervention ofthedisease state. Understanding the regulation between the Th1 and Th2 phenotype mediated by cytokines will help in designing cytokine-antagonist in therapeutics. The production ofIL-10 is strongly increased in mice infected with various pathogens such as Leishmania major, Schistosoma mansoni, Trypanosoma cruzi and Mycobacterium Leprae (Sher etal.,1992; Salgame etal., 1991, Heinzel etal., 1991).
When designing immune therapy to facilitate mounting the right arm ofdefense mechanism toward pathogens, it is important to maintain a balance between the two arms also. Th2-type responses may be important in controlling the tissue damage mediated by Thl cells during the response to an intracellular infectious agent. Keeping some Th1 cells functioning in a predominantly Th2 environment can help abrogate damaging effects of Thl by secreting IL-10 and IL-4. One extreme ofthespectrum ofThl/Th2 is reflected in transgenic mice lacking the IL-10 gene (Kuhn etal., 1993). The IL-10 deficient mouse is normal with respect to its development ofT and B cell subsets. However these mice develop chronic enterocolitis (or inflammatory bowel disease) due to chronic
inflammation via continuous overproduction ofcytokines such as TNFα and
IFN-gamma(Th1 response).
IL-12 can also inducethe development of theThl subset. By using Lysteria monocytogen, an intracellulargram-positive bacterium, infection inantibody T cell receptortransgenic mice as amodel ithas been shownthatIL-10 can block the production ofIL-12 from macrophages (Hsieh et al.,1993). Thus anIL-10-antagonistwill tip the Thl/Th2 populationpredominantlyto Th2 type environmentby 1, preventing the inhibition of the production of Thl cytokines 2 by allowingtheproductionofacytokine that induces the development ofThl subset.
With experimental evidence in hand it has been proposed that the resistance and/or progressionto AIDS is dependent on a Thl/Th2 stage ofan individual (Clerici & Shearer, 1993). This hypothesis is based onthe findings thatprogressionto AIDS is characterized by loss ofIL-2 and IFN-gammaproduction (loss ofThl response) with increase in IL-4 andIL-10 (acquired Th2 response). Many seronegatives (HIV-exposed individuals) generate a strong Thl-type response. It is importantto notethatafter seroconversionboth IL-4 and IL-10 levels go up at the expense of IL-4 and IFN-gamma. However, in full-blown AIDS patients, Th2 response seems to be mediated by high levels of IL-10 but not withIL-4, the level ofwhich goes downto normal inthese individuals. An anti-IL-10 reagentmay serve as apotentialtherapeutic in shiftingthe Th2 responseto Thl inAIDS patients to offer protection. TNFα
TNFα is an extracellularcytokine and acentral mediator of the immune and inflammatory response (Beutler et al., 1989; Vassalli, 1992). It is ahomo-trimer (Smith et al, 1987, Eck et al., 1988), andhas a subunit size of17 kD. It circulates at concentrations ofless than 5 pg/ml inhealthy individuals (Dinarello et al., 1993) and itcan go as high as 1000 pg/ml inpatients with sepsis syndrome (Casey et al., 1993). The human TNFα is nonglycosylated, whereas in some other species (notably the mouse) glycosylation occurs on a single N-linked site inthe mature protein, butthe sugarmoiety is not essential for biological activity (Beutler et al., 1989). Thehuman TNFα is acidic withapH of5.3 (Aggarwal et al., 1985). Each TNFα subunit consists ofan anti parallel β-sandwich and it participates in atrimer formation by an edge-to-face packing ofβ-sheets. The structure of theTNFα trimer resemblesthe "jelly-roll" structural motifcharacteristic ofviral coat
proteins (Jones etal., 1989). TNFα is a relatively stable molecule and may be exposed to chaotropic agents such as urea, SDS, or guanidinium hydrochloride, and renatured with recovery ofas much as 50% ofthe initial biological activity. The TNFα renaturability may reflect the limited number ofinternal disulfide bonds (one per monomer) required for maintenance ofstructure (Beutler etal., 1989).
Another related molecule, TNFβ, has the same bioactivity as TNFa. The interspecies sequence identity within the TNFα and TNFβ families is 71% and 61%, respectively (Beutler etal., 1989). The sequence identity between hTNF-α and hTNF-β is only 29% (Beutler etal., 1989). Despite their low similarity, both hTNFα and hTNFβ bind to the same receptors with comparable affinities.
TNFα mediates its bioactivity through binding to cell surface receptors. The TNFα receptors are found on the surface ofvirtually all somatic cells tested (Vassalli, 1992). Two distinct TNFα receptors have been characterized ofapparent molecular weights 55kD (p55 TNFα-Rl) and 75kD (p75 TNFα-R2) (Hohmann etal., 1989;
Brockhaus etal., 1990; Loetscheretal., 1991). Both receptors bind TNFα and TNFβ with high affinities (Kd=0.3-0.6 nM) (Loetscher etal., 1990; Schall etal., 1990; Pennicaetal., 1992).
TNFα has diverse activities, and thus is implicated in several diseases as follows:
Septic shock. Sepsis incidents have been increasing for the last 60 years and is the most common cause ofdeath in intensive care units in the United States (Parrillo, 1991).
The mortality ofseptic shock remains at approximately 50% despite the standard use of aggressive antibiotics and cardiovascular support for the past 10 years (Parrillo, 1991). The evidence implicating TNFα in sepsis is as follows. Pretreatment ofmice or baboons with monoclonal antibodies to TNFα protects them from lethal doses ofE. coli LPS (Beutler etal., 1985). Anti-TNFα antibodies protect primates against lethal endotoxin sepsis and against lethal S. aueus-induced shock (Fiedler etal., 1992; Hinshaw etal., 1992). Soluble-TNFα-receptor (p55)-IgG-Fc fusions (TNFα receptor immunoadhesin) were found to protect mice from endotoxic shock, even when administered lhr after endotoxin infusion. The same immunoadhesin was also effective against listeriosis in mice (Haak-Frendscho etal., 1994). Another immunoadhesin based on the p75 receptor
was also shown to be effective in lethal endotoxemia and it was functioning
simultaneously as both TNFα carrier and TNFα antagonist (Mohler etal., 1993).
Cachexia. In vivo administration ofTNFα causes cachexia in mice (Oliffetal., 1987). Therefore, TNFα antagonists may protect cancer or AIDS infected patients from cachexia.
Cerebral malaria. High levels ofTNFα are associated with poor prognosis in children with cerebral malaria, and antibodies to TNF a protect mice from cerebral complications ofPlasmodium berghei infection (Grau etal., 1987).
Arthritis. Antibodies to TNFα reduce the production ofthe inflammatory cytokine, IL-1 in synovial cells (Brennanetal., 1989). TNF α is an inducer of collagenase, the major destructive protease in rheumatoid arthritis (Brennan etal., 1989). Anti-TNF α antibodies were found to amelioratejoint disease in murine collagen-induced arthritis (Williams etal., 1992). Transgenic mice carrying the hTNFα gene develop arthritis which can be prevented by in vivo administration ofa monoclonal antibody against hTNFα (Kefferetal., 1991).
Graft Rejection and Graft versus Host Reaction (GVHR). TNFα has been implicated in the acute phase ofgraft-versus-host disease and in renal allograft rejection. Antagonists ofTNFα may then be able to prevent these life-threatening conditions.
Anti-TNFα antibodies have been found to delay graft rejection in experimental animals (Piguet, 1992). Also, injection ofanti-TNFα antibodies during the acute phase ofGVHR reduces mortality, and the severity ofintestinal, epidermal, and alveolar lesions (Piguet, 1992). Clinical trials ofthe efficacy ofanti-TNFα antibody in human bone marrow transplantation are underway.
AIDS. Studies ofintracellular signal transduction pathways revealed that TNFα induces proteins that bind to kB-like enhancer elements and thus takes part in the control ofNF-kB-inducible genes (Lenardo etal., 1989; Lowenthal etal., 1989; Osborn etal., 1989). The antiviral activity ofTNFα at least in part is mediated by the interaction of NF-kB with a virus-inducible element in the β-interferon gene (Goldfeld etal., 1989; Visvanathan etal., 1989). By an analogous mechanism, TNFα appears to activate human immunodeficiency virus type I (Duh etal., 1989; Folks etal., 1989). Therefore, TNFα
antagonists may prove useful in delaying the activation ofthe AIDS virus and may work in conjunction with other treatments in the cure ofAIDS.
Parkinson's disease. Recently, elevated TNFα levels have been found in the brain and the cerebrospinal fluid ofParkinsonian patients (Mogi etal., 1994). This report speculates that elevated TNFα levels may be related to neuronal degeneration associated with the disease.
RANTES
RANTES is a small (MW 8-kD) highly basic (pl~9.5) chemokine that belongs to the CC group (Schall, 1991; Baggiolini etal., 1994). It does not appear to be glycosylated (Schall, 1991) and is a chemoattractant for monocytes (Schall etal., 1990; Wang etal., 1993; Wiedermann etal., 1993), basophils (Bischoffet al, 1993; Kunaetal., 1993), eosinophils (Rotetal., 1992), and CD4+/UCHL1+T lymphocytes which are thought to be prestimulated or primed helper T cells involved in memory T cell function (Schall etal., 1990). RANTES is not only a chemoattractant but it also stimulates cells to release their effectors leading to tissue damage. For example, RANTES causes histamine release from basophils (Kunaetal., 1992; Kunaetal., 1993; Alam etal., 1993). It also causes the secretion ofeosinophil basic peptide (Alam etal., 1993) and the production ofoxygen free radicals (Rot etal., 1992) by eosinophils.
Initially, it was thought that RANTES was synthesized by activated T cells but recently other cells were found to synthesize it very fast upon stimulation. RANTES mRNA is expressed late (3 to 5 days) after activation ofresting T cells, whereas in fibroblasts, renal epithelial and mesangial cells, RANTES mRNA is quickly up-regulated by TNFα stimulation (Nelson etal., 1993).
Receptors for RANTES have been identified. There is a promiscuous receptor on the surface oferythrocytes that binds all chemokines with a Kd=5nM (Horuk etal., 1993; Neote etal., 1993). This receptor is thought to be a sink for chemokines to help in the establishment ofchemotactic gradients. Signal transducing receptors have also been identified and cloned (Gao etal., 1993; Neote etal., 1993; Van-Riper etal., 1993; Wang etal., 1993). Monocytes carry a G-protein coupled receptor that binds RANTES with estimated Kd of400 pM, but also MCAF and MlP-la with lower affinities (estimated Kd
of6 and 1.6 nM respectively) (Wang et al., 1993). Areceptormolecule has been cloned from neutrophils that canbind RANTES with a lower affinity ofabout 50 nM (Gao et al., 1993).
Disease State. RANTES antagonists may have therapeutic application in inflammation. Blockage ofthe chemoattractant and effector cell activationproperties of RANTES would block local inflammationandtissue damage. The mechanism ofaction ofthe RANTES antagonistwill bethe inhibition ofRANTES bindingto cell surface receptors.
RANTES is chemoattractant formonocytes, basophils, eosinophils and memory lymphocytes. Basophils are the major source ofmediators such as histamine and peptido-leukotrienes, and are an essential element ofthe late-phase responses to allergens inhypersensitivity diseases. These cells are also involved in otherinflammatory pathologies, including certain autoimmune reactions, parasitic infections and inflammatory bowel diseases. Inthese conditions, basophil recruitement and activation is independent ofIgE. Numerous reports have accumulated overthe years that describe the effects ofa group ofelusive stimuli operationally called "histamine-releasing factors." A large number ofthese elusive stimuli may well be contributedby RANTES.
Eosinophiles also are important in allergic inflamation, andtogetherwith lymphocytes, form prominent infiltrates inthe bronchial mucosa ofpatients with asthma. They are believedto be the cause ofepithelial damage and the characteristic airway hyper-reactivity. The recruitement of lymphocytes of the Th2 type, which comigrate with eosinophiles into sites oflate-phasereactions, is animportant source ofother
chemoattractant cytokines and growthfactors thatprime eosinophils.
RANTES, with its effects onmonocytes, basophils, eosinophils and lymphocytes appears to be apotent stimulatorofeffector-cell accumulation and activation in chronic inflammatory diseases and inparticular, allergic inflammation.
The recruitement system ofinflammatory cells has some redundancy built into it. However, RANTES has some unique properties. It is a more potent chemoattractant than MCP-1 and MIP-1 a , while MCP-1 is more potent stimulatorofhistamine release from basophils (Baggiolini et al., 1994). RANTES causes the production ofoxygen radicals by eosinophiles while MIP-1 a cannot (Rot et al., 1992). RANTES is as potent as C5a in the
recruitement ofeosinphiles, but not as potent a trigger ofthe eosinophil oxydation burst (Rot et al., 1992). C5a is a very potent chemoattractant: however, it lacks the specificity ofRANTES. It attracts not only basophils and eosinophils but also neutrophils. Since the eosinophils, but not the neutrophils, are important in the pathophysiology ofsome inflammatory conditions, such as the allergen-induced late-phase reaction and asthma, specific chemoattractants such as RANTES are expected to be involved.
Using in situ hybridization, RANTES expression has been found in interstitial mononuclear cells and proximal tubular epithelial cells in human kidney transplants undergoing rejection. Antibody staining revealed the presence ofRANTES not only within the interstitial infiltrate and renal tubular epithelial cells but also in high abundance in inflamed endothelium (Wiedermann etal., 1993). Based on these results a haptotactic mechanism was postulated. Haptotaxis is defined as cell migration induced by
surface-bound gradients. The haptotactic mechanism was supported by in vitro experiments and anti-RANTES antibodies have been found to prevent that in vitro haptotaxis.
Human rheumatoid synovial fibroblasts express mRNA for RANTES and IL-8 after stimulation with TNFa and IL-lβ (Rathanaswami etal., 1993). There is a differential regulation ofexpression ofIL-8 and RANTES mRNA. Cycloheximide enhanced the mRNA levels for IL-8 and RANTES after stimulation with IL-lβ but reduced the levels ofRANTES mRNA after stimulation with TNFα. Also, IL-4 down-regulates and IFN-gamma enhances the TNFα and IL-lβ induced increase in RANTES mRNA, whereas the induction ofIL-8 mRNA by TNFα or IL-lβ was inhibited by IFN-gamma and augmented by IL-4. Moreover, the combination ofTNFα and IL-lβ synergistically increased the level ofIL-8 mRNA, whereas under the same conditions, the levels ofRANTES mRNA were less than those induced with TNFα alone. These studies suggest that the synovial fibroblasts may participate in the ongoing inflammatory process in rheumatoid arthritis, and RANTES might be one ofthe participating effectors. The observed differential regulation ofIL-8 and RANTES indicates that the type ofcellular infiltrate and the progress ofthe inflammatory disease is likely to depend on the relative levels ofstimulatory and inhibitory cytokines.
RANTES has also been implicated in atherosclerosis and possibly in
postangioplasty restenosis (Schall, 1991). The participation ofMCP-1 in atherosclerosis hasbeen studiedto agreaterextent. RecentlymRNAs forRANTES, MIP-1 a and MP-1β have been detected in in situ in normal carotidplaque and heart transplant atherosclerosis. RANTES mRNA is not detected inthe same cells expressing MlP-la and MlP-lβ, but it is expressed in lymphocytes andmacrophages typically more proximal to the lumen. The dataargue forpositive feed-backmechanisms forthe CC chemokines and possible differential expressionofthese chemokines atvarious stages inthe progression of arterial disease.
Finally, elevated RANTES levels have been correlated with endometriosis
(Khorram etal, 1993). RANTES levels were elevated inpelvic fluids fromwomen with endometriosis, andthese levels correlate withthe severity ofthe disease.
Protein HomologybetweenHuman andAnimal. The murine RANTES has been cloned (Schall etal, 1992). Sequence analysis revealed 85% amino acid identity between the human andmouseproteins. The humanandmurine RANTES exhibit immune crossreactivity. Boyden chamberchemotaxis experiments reveal some lack ofspecies specificity inmonocyte chemoattractantpotential, as recombinant muRANTES attracts humanmonocytes in adose-dependent fashion in vitro. Also, hRANTES transfection into mouse tumorcell lines produce tumors in which the secretion of hRANTES by those tumors correlates with increased murine monocyte infiltration in vivo (Schall et al., 1992).
SELEX
A method forthe in vitro evolution ofnucleic acid molecules with highly specific binding to target molecules has been developed. This method, Systematic Evolution of Ligands by Exponential enrichment, termed SELEX, is described in United States Patent Application Serial No. 07/536,428, entitled "Systematic Evolution of Ligands by
Exponential Enrichment," nowabandoned, United States Patent Application Serial No. 07/714,131, filed June 10, 1991, entitled "Nucleic Acid Ligands," United States Patent Application Serial No.07/931,473, filed August 17, 1992, entitled "Nucleic Acid
Ligands," nowUnited States PatentNo.5,270,163 (see also PCT/US91/04078), each of which is herein specifically incorporated by reference. Each of these applications.
collectively referred to herein as the SELEX Patent Applications, describes a
fundamentally novel method for making a nucleic acid ligand to any desired target molecule.
The SELEX method involves selection from a mixture ofcandidate
oligonucleotides and step-wise iterations ofbinding, partitioning and amplification, using the same general selection scheme, to achieve virtually any desired criterion ofbinding affinity and selectivity. Starting from a mixture ofnucleic acids, preferably comprising a segment ofrandomized sequence, the SELEX method includes steps ofcontacting the mixture with the target under conditions favorable for binding, partitioning unbound nucleic acids from those nucleic acids which have bound specifically to target molecules, dissociating the nucleic acid-target complexes, amplifying the nucleic acids dissociated from the nucleic acid-target complexes to yield a ligand-enriched mixture ofnucleic acids, then reiterating the steps ofbinding, partitioning, dissociating and amplifying through as many cycles as desired to yield highly specific, high affinity nucleic acid ligands to the target molecule.
The basic SELEX method has been modified to achieve a number ofspecific objectives. For example, United States Patent Application Serial No.07/960,093, filed October 14, 1992, entitled "Method for Selecting Nucleic Acids on the Basis ofStructure," describes the use of SELEX in conjunction with gel electrophoresis to select nucleic acid molecules with specific structural characteristics, such as bent DNA. United States Patent Application Serial No.08/123,935, filed September 17, 1993, entitled "Photoselection of Nucleic Acid Ligands" describes a SELEX based method for selecting nucleic acid ligands containing photoreactive groups capable ofbinding and/or photocrosslinking to and/or photoinactivating a target molecule. United States Patent Application Serial No.
08/134,028, filed October 7, 1993, entitled "High-Affinity Nucleic Acid Ligands That
Discriminate Between Theophylline and Caffeine," describes a method for identifying highly specific nucleic acid ligands able to discriminate between closely related molecules, termed Counter-SELEX. United States Patent Application Serial No.08/143,564, filed October 25, 1993, entitled "Systematic Evolution ofLigands by Exponential Enrichment: Solution SELEX," describes a SELEX-based method which achieves highly efficient partitioning between oligonucleotides having high and low affinity for a target molecule.
United States Patent Application Serial No. 07/964,624, filed October 21, 1992, entitled "Methods ofProducingNucleic Acid Ligands" describes methods forobtaining improved nucleic acid ligands after SELEX has beenperformed. United States Patent Application SerialNo.08/400,440, filed March 8, 1995, entitled "Systematic.Evolution of Ligands by Exponential Enrichment: Chemi-SELEX," describes methods forcovalently linking a ligandto itstarget.
The SELEX method encompasses the identification of high-affinity nucleic acid ligands containing modified nucleotides conferring improved characteristics on the ligand, such as improved in vivo stability or improved delivery characteristics. Examples of such modifications include chemical substitutions atthe ribose and/orphosphate and/orbase positions. SELEX-identified nucleic acid ligands containing modified nucleotides are described inUnited States Patent Application SerialNo.08/117,991, filed September 8, 1993, entitled "High Affinity Nucleic Acid Ligands Containing Modified Nucleotides," that describes oligonucleotides containing nucleotide derivatives chemically modified at the 5- and 2'-positions of pyrimidines. United States Patent Application Serial No.
08/134,028, supra, describes highly specific nucleic acid ligands containing one ormore nucleotides modified with 2'-amino (2'-NH2), 2'-fluoro (2'-F), and/or2'-O-methyl
(2'-OMe). United States Patent Application Serial No.08/264,029, filed June 22, 1994, entitled "Novel Method of Preparation of 2' Modified Pyrimidine Intramolecular
Nucleophilic Displacement," describes oligonucleotides containing various 2'-modified pyrimidines.
The SELEXmethod encompasses combining selected oligonucleotides with other selected oligonucleotides and non-oligonucleotide functional units as described inUnited States PatentApplication Serial No.08/284,063, filed August 2, 1994, entitled
"Systematic Evolution ofLigands by Exponential Enrichment: Chimeric SELEX" and United States Patent Application Serial No.08/234,997, filed April 28, 1994, entitled "Systematic Evolution ofLigands by Exponential Enrichment: Blended SELEX," respectively. These applications allowthe combination ofthe broad array ofshapes and other properties, and the efficient amplification and replication properties, of
oligonucleotides with the desirable properties ofother molecules. Each of the above
described patent applications which describe modifications ofthe basic SELEX procedure are specifically incorporated by reference herein in their entirety.
BRIEF SUMMARY OF THE INVENTION
The present invention includes methods ofidentifying and producing nucleic acid ligands to cytokines and the nucleic acid ligands so identified and produced. In particular, RNA sequences are provided that are capable ofbinding specifically to IFN-gamma, IL-4, IL-10, and TNFα. In addition, DNA sequences are provided that are capable ofbinding specifically to RANTES. Specifically included in the invention are the RNA ligand . sequences shown in Tables 3, 4, 7, 8, 10, and 12 (SEQ ID NOS:7-73; 79-185; 189-205; 209-255).
Further included in this invention is a method ofidentifying nucleic acid ligands and nucleic acid ligand sequences to a cytokine comprising the steps of(a) preparing a candidate mixture ofnucleic acids, (b) contacting the candidate mixture ofnucleic acids with a cytokine, (c) partitioning between members ofsaid candidate mixture on the basis ofaffinity to the cytokine, and (d) amplifying the selected molecules to yield a mixture of nucleic acids enriched for nucleic acid sequences with a relatively higher affinity for binding to the cytokine.
Further included in this invention is a method ofidentifying nucleic acid ligands and nucleic acid ligand sequences to a cytokine selected from the group consisting of IFN-gamma, IL-4, IL-10, TNFα, and RANTES comprising the steps of(a) preparing a candidate mixture ofnucleic acids, (b) contacting the candidate mixture ofnucleic acids with said cytokine, (c) partitioning between members ofsaid candidate mixture on the basis ofaffinity to said cytokine, and (d) amplifying the selected molecules to yield a mixture ofnucleic acids enriched for nucleic acid sequences with a relatively higher affinity for binding to said cytokine.
More specifically, the present invention includes the RNA ligands to IFN-gamma, IL-4, IL-10, and TNFα identified according to the above-described method, including those ligands shown in Tables 3, 4, 7, 8, 10, and 12 (SEQ ID NOS:7-73; 79-185; 189-205; 209-255). Also included are RNA ligands to IFN-gamma, IL-4, IL-10, and TNFα that are substantially homologous to any ofthe given ligands and that have substantially the same
ability to bind IFN-gamma, IL-4, IL-10, and TNFα and inhibit the function of
IFN-gamma, IL-4, IL-10, and TNFα. Furtherincludedinthis invention are nucleic acid ligands to IFN-gamma, IL-4, IL-10, and TNFα thathave substantially the same structural form as the ligands presented herein and thathave substantially the same ability to bind IFN-gamma, IL-4, IL-10, and TNFα and inhibit the function of IFN-gamma, IL-4, IL-10, andTNFα.
The present invention also includes modified nucleotide sequences based on the nucleic acidligands identified herein andmixtures of the same. DETAILED DESCRIPTION OF THE INVENTION
This application describes high-affinity nucleic acid ligands to cytokines identified through the method known as SELEX. SELEXis described inU.S. PatentApplication SerialNo.07/536,428, entitled SystematicEvolution of Ligands byExponential
Enrichment, nowabandoned, U.S. PatentApplication Serial No.07/714,131, filed June 10, 1991, entitledNucleic Acid Ligands, United States Patent Application Serial No. 07/931,473, filed August 17, 1992, entitled Nucleic Acid Ligands, now United States PatentNo.5,270,163, (see also PCT/US91/04078). These applications, each specifically incorporated herein by reference, are collectively called the SELEX Patent Applications.
Inits mostbasic form, the SELEXprocess maybe defined by the following series ofsteps:
1) A candidate mixture ofnucleic acids ofdiffering sequence is prepared. The candidate mixture generally includes regions offixed sequences (i.e., each of the members of the candidate mixture contains the same sequences inthe same location) andregions of randomized sequences. The fixed sequenceregions are selected either: (a) to assist inthe amplification steps describedbelow, (b) to mimic a sequence knownto bind to the target, or(c) to enhance the concentration ofa given structural arrangement ofthe nucleic acids in the candidate mixture. The randomized sequences can be totally randomized (i.e., the probability of finding a base at any position being one in four) oronly partially
randomized (e.g., theprobability offinding a base at any location can be selected at any level between 0 and 100 percent).
2) The candidate mixture is contacted with the selected target under conditions favorable for binding between the target and members ofthe candidate mixture. Under these circumstances, the interaction between the target and the nucleic acids ofthe candidate mixture can be considered as forming nucleic acid-target pairs between the target and those nucleic acids having the strongest affinity for the target.
3) The nucleic acids with the highest affinity for the target are partitioned from those nucleic acids with lesser affinity to the target. Because only an extremely small number ofsequences (and possibly only one molecule ofnucleic acid) corresponding to the highest affinity nucleic acids exist in the candidate mixture, it is generally desirable to set the partitioning criteria so that a significant amount ofthe nucleic acids in the candidate mixture (approximately 5-50%) are retained during partitioning.
4) Those nucleic acids selected during partitioning as having the relatively higher affinity to the target are then amplified to create a new candidate mixture that is enriched in nucleic acids having a relatively higher affinity for the target.
5) By repeating the partitioning and amplifying steps above, the newly formed candidate mixture contains fewer and fewer weakly binding sequences, and the average degree ofaffinity ofthe nucleic acids to the target will generally increase. Taken to its extreme, the SELEX process will yield a candidate mixture containing one or a small number ofunique nucleic acids representing those nucleic acids from the original candidate mixture having the highest affinity to the target molecule.
The SELEX Patent Applications describe and elaborate on this process in great detail. Included are targets that can be used in the process; methods for partitioning nucleic acids within a candidate mixture; and methods for amplifying partitioned nucleic acids to generate enriched candidate mixture. The SELEX Patent Applications also describe ligands obtained to a number oftarget species, including both protein targets where the protein is and is not a nucleic acid binding protein.
The nucleic acid ligands described herein can be complexed with a lipophilic compound (e.g., cholesterol) or attached to or encapsulated in a complex comprised of lipophilic components (e.g., a liposome). The complexed nucleic acid ligands can enhance the cellular uptake ofthe nucleic acid ligands by a cell for delivery ofthe nucleic acid ligands to an intracellular target U.S. Patent Application No.08/434,465, filed May 4,
1995, entitled "Nucleic Acid Ligand Complexes," which is incorporated in its entirety herein, describes amethodforpreparing atherapeutic or diagnostic complex comprised of anucleic acid ligand and a lipophilic compound or anon-immunogenic, high molecular weight compound.
The methods described herein andthe nucleic acid ligands identified by such methods are useful for both therapeutic and diagnostic purposes. Therapeutic uses include thetreatment orpreventionofdiseases ormedical conditions inhumanpatients.
Diagnostic utilizationmay include both in vivo or in vitro diagnostic applications. The SELEX method generally, and the specific adaptations of the SELEX method taught and claimed herein specifically, areparticularly suited fordiagnostic applications. SELEX identifies nucleic acid ligands thatare able to bindtargets with high affinity and with surprising specificity. These characteristics are, ofcourse, the desiredproperties one skilled in the art would seek in a diagnostic ligand.
The nucleic acid ligands of the present invention may be routinely adapted for diagnosticpurposes according to any numberoftechniques employed by those skilled in the art. Diagnostic agents need only be able to allowthe userto identify thepresence ofa giventarget at aparticular locale orconcentration. Simplythe ability to form binding pairs withthe targetmaybe sufficienttotriggerapositive signal fordiagnostic purposes. Those skilled inthe artwould also be ableto adapt any nucleic acid ligand byprocedures known in the art to incorporate a labeling tag in order to track the presence of such ligand. Such atag could be used in anumber ofdiagnostic procedures. The nucleic acid ligands to cytokines described herein may specifically be used for identification of the cytokine proteins.
SELEX provides high affinity ligands of a target molecule. This represents a singularachievementthat is unprecedented inthe field ofnucleic acids research. The present invention applies the SELEX procedure to the specific target. Inthe Example section below, the experimental parameters used to isolate and identify the nucleic acid ligands to cytokines are described.
In orderto produce nucleic acids desirable foruse as apharmaceutical, it is preferred that the nucleic acid ligand (1) binds to the target in a mannercapable of achieving the desired effect on the target; (2) be as small as possible to obtain the desired
effect; (3) be as stable as possible; and (4) be a specific ligand to the chosen target. In most situations, it is preferred that the nucleic acid ligand have the highest possible affinity to the target.
In co-pending and commonly assigned U.S. Patent Application Serial No.
07/964,624, filed October 21, 1992 ('624), methods are described for obtaining improved nucleic acid ligands after SELEX has been performed. The '624 application, entitled Methods ofProducing Nucleic Acid Ligands, is specifically incorporated herein by reference.
This invention includes the SELEX process for identification ofnucleic acid ligands ofcytokines. Cytokines are a diverse group ofsmall proteins that mediate cell signaling/communication. Cytokines include immune/ hematopoietins (e.g.,EPO, GM-CSF, G-CSF, LIF, OSM, CNTF, GH, PRL, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-9, IL-10, IL-12), interferons (e.g.,IFNα, IFNβ, IFN-gamma), TNF-related molecules
(e.g.,TNFα, IFNβ, gp39 (CD40-L), CD27-L, CD30-L,NGF), and chemokines (e.g, PF4, PBP, groα, MIG, ENA-78, MlPlα, MIPl β, MCP-1, 1-309, HC14, CIO, RANTES, IL-8, MIP-1). In one embodiment, cytokines are derived from T-lymphocytes.
In the present invention, SELEX experiments were performed in order to identify RNA with specific high affinity for the cytokines IFN-gamma, IL-4, IL-10, hTNFα, and RANTES from degenerate libraries containing 30 or 40 random positions (40N for IFN-gamma, IL-4, IL-10 and RANTES; 30N for hTNFα) (Tables 1, 5, 9, 11, and 16). This invention includes the specific RNA ligands to IFN-gamma, IL-4, IL-10, and TNFα shown in Tables 3, 4, 7, 8, 10, and 12 (SEQ ID NOS:7-73; 79-185; 189-205; 209-255), identified by the methods described in Examples 1, 3, 5, 7, and 12. This invention further includes RNA ligands to IFN-gamma, IL-4, IL-10, and TNFα which inhibit the function of IFN-gamma, IL-4, IL-10, and TNFα. This invention further includes DNA ligands to
RANTES which inhibit the function ofRANTES. The scope ofthe ligands covered by this invention extends to all nucleic acid ligands ofIFN-gamma, IL-4, IL-10, TNF a, and RANTES modified and unmodified, identified according to the SELEX procedure. More specifically, this invention includes nucleic acid sequences that are substantially homologous to the ligands shown in Tables 3, 4, 7, 8, 10, and 12 (SEQ ID NOS:7-73;
79-185; 189-205; 209-255). By substantially homologous it is meant a degree ofprimary
sequence homology in excess of 70%, most preferably in excess of 80%. A review of the sequence homologies of the ligands of IFN-gamma, IL-4, IL-10, and TNFα shown in Tables 3, 4, 7, 8, 10, and 12 (SEQ IDNOS:7-73; 79-185; 189-205; 209-255) shows that sequences with little orno primary homology may have substantially the same ability to bindIFN-gamma, IL-4, IL-10, and TNFα. Forthese reasons, this invention also includes nucleic acid ligands thathave substantially the same structure and abilityto bind
IFN-gamma, IL-4, IL-10, and TNFα asthe nucleic acid ligands shown in Tables 3, 4, 7, 8, 10, and 12 (SEQ IDNOS:7-73; 79-185; 189-205; 209-255). Substantiallythe same ability to bindIFN-gamma, IL-4, IL-10, and TNFα means thatthe affinity is within one ortwo orders ofmagnitude ofthe affinity of the ligands described herein. It is well withinthe skill ofthose ofordinary skill in the artto determine whether agiven sequence— substantiallyhomologous to those specificallydescribedherein— has substantially the same ability to bind IFN-gamma, IL-4, IL-10, and TNFα.
This inventionalso includes the ligands as described above, wherein certain chemical modifications are made in orderto increase the in vivo stability of the ligand or to enhance ormediate the delivery ofthe ligand. Examples of such modifications include chemical substitutions at the sugarand/orphosphate and/or base positions ofagiven nucleic acid sequence. See, e.g., U.S. Patent Application Serial No. 08/117,991, filed September 9, 1993, entitled High Affinity Nucleic Acid Ligands Containing Modified Nucleotides whichis specifically incorporatedhereinbyreference. Othermodifications are knownto one ofordinary skill in the art. Suchmodifications may be made
post-SELEX (modification ofpreviously identified unmodified ligands) orby
incorporation into the SELEXprocess.
As described above, because oftheirabilityto selectively bind IFN-gamma, IL-4, IL-10, hTNFα, and RANTES,the nucleic acid ligandsto IFN-gamma, IL-4, IL-10,TNFα, and RANTES described herein are useful as pharmaceuticals. This invention, therefore, also includes a method ofinhibiting cytokine function by administration ofa nucleic acid ligand capable ofbinding to acytokine.
Therapeutic compositions of the nucleic acid ligands may be administered parenterally by injection, although other effective administration forms, such as intraarticular injection, inhalant mists, orally active formulations, transdermal
iontophoresis or suppositories, are also envisioned. One preferred carrier is physiological saline solution, but it is contemplated that other pharmaceutically acceptable carriers may also be used. In one preferred embodiment, it is envisioned that the carrier and the ligand constitute a physiologically-compatible, slow release formulation. The primary solvent in such a carrier may be either aqueous or non-aqueous in nature. In addition, the carrier may contain other pharmacologically-acceptable excipients for modifying or maintaining the pH, osmolarity, viscosity, clarity, color, sterility, stability, rate ofdissolution, or odor of the formulation. Similarly, the carrier may contain still other
pharmacologically-acceptable excipients for modifying or maintaining the stability, rate of dissolution, release, or absorption ofthe ligand. Such excipients are those substances usually and customarily employed to formulate dosages for parental administration in either unit dose or multi-dose form.
Once the therapeutic composition has been formulated, it may be stored in sterile vials as a solution, suspension, gel, emulsion, solid, or dehydrated or lyophilized powder. Such formulations may be stored either in a ready to use form or requiring reconstitution immediately prior to administration. The manner ofadministering formulations containing nucleic acid ligands for systemic delivery may be via subcutaneous,
intramuscular, intravenous, intranasal or vaginal or rectal suppository.
The following Examples are provided to explain and illustrate the present invention and are not intended to be limiting ofthe invention.
EXAMPLE 1. EXPERIMENTAL PROCEDURES FOR 2-NH2AND 2'- F-MODIFIED LIGANDS TO IFN-GAMMA
This example provides general procedures followed and incorporated in Example 2 for the evolution ofnucleic acid ligands to IFN-gamma.
A. Oligonucleotides
2'F modified CTP and UTP were prepared according to the method ofPieken et al, 1991. 2'NH2 modified CTP and UTP were prepared according to the method ofMcGee et al, U.S. Patent Application No.08/264,029, filed June 22, 1994, which is incorporated herein by reference (see also McGee et al.1995). DNA oligonucleotides were
synthesized by Operon Technologies (Alameda CA).
B. SELEX
The SELEXprocedure has been described in detail in U.S. PatentNo. 5,270,163 (see also Tuerk and Gold, 1990; Gold et al., 1993). Three SELEXprocedures were performed to evolve high affinity ligandsto IFN-gamma. Each SELEX procedure utilized RNApools containingpyrimidines modified at the 2' position as follows, 1) 2'F-CTP and 2'F-UTP referred to as 2'F, 2) 2'F-CTP and 2-NH2-UTP referred to as 2'F/NH2, and 3) 2;NH2-CTP and 2'NH2-UTP referredto as 2'NH2. For each SELEX, the DNA template 40N7 was designedto contain 40 randomnucleotides, flanked by 5' and 3' regions of fixed sequence (Table 1; SEQ ID NO:1). The fixed regions include DNA primer annealing sites forPCRand cDNA synthesis as well as the consensus T7 promoter region to allow in vitro transcription.
Single-stranded DNAprimers andtemplates were synthesized and amplified into double-strandedtranscribable templates by PCR. Preparation ofthe initial pool ofRNA molecules involvedPCR amplification of1000 pmoles ofsingle-stranded template (Table 1; SEQ ID NO:l) and 2500 pmoles of both the 5'- (5P7; SEQ ID NO:2) and 3'- (3P7; SEQ ID NO:3) primers. These were incubated in a reaction mixture containing 50 mM KCl , 10 mM Tris-Cl (pH 8.3), 3 mM MgCl2, 0.5 mM of each dATP, dCTP, dGTP, and dTTP. Taq DNA Polymerase (Perkin-Elmer, Foster City CA) at 0.1 U/μl was added andthe reaction incubated at 97°C for3 minto denaturethetemplate andprimers. Following the initial denaturing step, the reactionwas cycled 10 times at 93°C for 30 sec, 53oC for 30 sec, and 72°C for 1 minto denature, anneal, and extend, respectively, theprimers and template. To get an accurate concentration ofdouble-stranded PCRproduct forthe initial round of SELEX, the PCRproductwas purifiedusing QIAquick-spin PCR purification columns (QIAGEN Inc., Chatsworth CA) as specified by the manufacturer.
For in vitro transcriptionusing modified nucleotides 200 pmoles (final
concentration of1 μM) of double-stranded DNA template was incubated in areaction mixture containing 40 mM Tris-Cl (pH 8.0), 12 mM MgCl2, 5 mM DTT, 1 mM
spermidine, 0.002% TritonX-100, 4% PEG 8000, 0.5 μM α-32P-ATP, 5 U/μl T7 RNA Polymerase (Davanloo et al., 1984), and concentrations ofothernucleotides as follows, 1) forthe 2'FSELEX: 1 mM ATP and GTP, 3 mM 2'F-CTP and 2'F-UTP, 2) for the
2'F/NH2 SELEX: 1 mM ATP, GTP, and 2'NH,-UTP and 3 mM 2'F-CTP, and 3) for the
2'NH2 SELEX: 1 mM ATP, GTP, 2'NH2-CTP, and 2'NH2-UTP. These incubations were performed in a 37°C incubator for between 6 hrs and overnight. Typically the RNA was purified by gel purification and elution. To expedite the process, for rounds 11, 12, and 14-17 the RNA was purified using Bio-Spin 6 chromatography columns (Bio-Rad Laboratories, Hercules CA) according to manufacturer's specifications. To reduce background, the RNA was pre-filtered prior to all rounds of SELEX except rounds 1, 2, 4, 6, 14, and 16. The pre-filtration step involved bringing the RNA up to 200 μl in phosphate buffered saline (PBS), modified to contain ImM Mg2+ ions, (138 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4, 1.1 mM KH2PO4, ImM MgCl2, pH 7.4), (mPBS), and passing this RNA solution through three filter discs (0.45 μm, nitrocellulose/ cellulose acetate, Millipore Corporation, Bedford MA) pre-wetted with mPBS.
For initial binding, 1000 pmoles ofRNA were incubated with human IFN-gamma protein in binding buffer, (mPBS plus 0.01% human serum albumin (HSA)), for 5-10 min at 37°C to allow binding to occur. Human recombinant IFN-gamma used in this SELEX procedure was purchased from two different sources. The first three rounds ofboth the 2'F and 2'F/NH2 SELEX were performed with protein obtained from Upstate Biotechnology, Lake Placid NY. The subsequent rounds ofthese two SELEX procedures as well as the entire 2'NH2 SELEX were performed with protein obtained from Genzyme Inc.,
Cambridge MA. For each round ofSELEX the concentration ofRNA and protein was carefully chosen to provide optimum stringency. Increased stringency was obtained during rounds 8-13 ofSELEX by adding NaCl to the binding buffer to bring the final chloride ion concentration up to 250 mM. Preliminary experiments had shown that IFN-gamma had a tendency to aggregate at high protein concentrations. To prevent the evolution ofRNA species having an affinity for this aggregated IFN-gamma, beginning with round 4 ofSELEX and for all subsequent rounds ofthe SELEX procedure, the binding mix was centrifuged at 16,000 x g for 3 min in an eppendorfcentrifuge before nitrocellulose filter partitioning. IFN-gamma/RNA complexes were separated from unbound RNA by nitrocellulose filter partitioning described below.
For nitrocellulose partitioning, the 2'F and 2'F/NH2 SELEX procedures used 0.2 μm pore size pure nitrocellulose filters (Scleicher & Schuell, Keene NH) for the first two rounds of SELEX. All subsequent rounds ofthese two SELEX procedures and the entire
2'NH2 SELEX were performed with 0.45 μm pore size nitrocellulose/cellulose acetate mixed matrix filters (Millipore Corporation, BedfordMA). Filterdiscswereplaced into a vacuum manifold andwetted with 5 ml ofmPBS buffer. The IFN-gamma/RNA binding mix was aspiratedthroughthe filter discs whichwere immediately washed with 5 ml of mPBS buffer. To fu rther increase stringency andreduce background for rounds 8-13, and 15, this washing step was modifiedto include washing of the filter discs with 15 ml 0.5 M ureafollowedby 20 ml mPBS buffer. Bound RNAwas isolated from filters by extraction in a solution of400 m1 phenol (equilibrated in Tris-Cl, pH 8.0)/300 m 17 M urea (freshly prepared). The filters were bathed inthe phenol/urea solution atroomtemperature for 30 min and at 95°C for2 min. The RNAwas phenol/chloroform extracted and ethanol precipitated with20 mg tRNA.
The RNAwas reversetranscribed into cDNA by addition of 50 pmoles DNA primer, 0.4 mMeachofdNTPs, and 1 U/μlAMVreversetranscriptase (AMV RT) (Life Sciences, Inc., St. Petersburg FL) in buffer containing 50 mM Tris-Cl (pH 8.3), 60 mM NaCl, 6 mM Mg(OAc)2, 10 mM DTT. The reactionwas incubated at 37°C for 30 min then 48°C for 30 minthen 70°C for 10 min, to ensure the melting ofsecondary structure present inthe isolated RNA.
To begin a new round of SELEX, the cDNA was PCR amplified by addition of 250 pmoles of both the 5' (5P7; SEQ ID NO:2) and 3' (3P7; SEQ ID NO:3) primer in reaction conditions identical to those detailed above. The numberofcycles of PCR required to amplifythe cDNA was carefully calculated for eachround of SELEX so that 250 pmoles double-stranded DNA template would be used to initiate the next round of SELEX.
C. Equilibrium Dissociation Constants (Kds)
The determination ofequilibrium dissociation constants (Kds) for RNA pools was made subsequentto rounds 5, 8, 12, and 17 to monitor the progress of each SELEX . The Kds of RNA pools formouse IFN-gamma (Genzyme Inc., Cambridge MA) were also determined after rounds 8 and 17. Kds were determined for individual ligands after cloning and sequencing ofRNA pools and truncations (described below). Nitrocellulose filter binding was used to determine Kds as follows: filter discs were placed into a vacuum manifold and wetted with 5 ml of mPBS buffer. "P-labeled-RNA was incubated
with serial dilutions ofIFN-gamma in binding buffer for 5-10 min at 37°C to allow binding to occur. Binding mixes were centrifuged as described above to remove aggregates, aspirated through the filter discs, and then immediately washed with 5 ml mPBS buffer. The filter discs were dried and counted in a liquid.scintillation counter (Beckmann Instruments, Palo Alto CA). Equilibrium dissociation constants were determined by least square fitting ofthe data points using the Kaleidagraph™ graphics program (Synergy Software, Reading PA). Many ligands and evolved RNA pools yield biphasic binding curves. Biphasic binding can be described as the binding oftwo affinity species that are not in equilibrium. Biphasic binding constants were calculated according to standard procedures. Kds were determined by least square fitting ofthe data points using the Kaleidagraph™ graphics program.
D. Cloning and Sequencing
After the 17th round of SELEX, RNA molecules were reverse transcribed to cDNA and made double-stranded by PCR amplification with primers containing recognition sites for the restriction endonucleases Hind III (Table 1; 5' primer 5P7H; SEQ ID NO:4) and Bam HI (Table 1; 3' primer 3P7B; SEQ ID NO:5). Using these restriction sites the DNA sequences were inserted directionally into the pUC19 vector. These recombinant plasmids were transformed into Epicurian coli JM109 competent cells (Stratagene, La Jolla CA). Plasmid DNA was prepared with the PERFECTprep™ plasmid DNA kit (5 ρrime->3 prime, Boulder CO). Plasmid clones were sequenced using a PCR sequencing protocol (Adams etal., 1991) using PCR sequencing primer pUC19F30 (SEQ ID NO:6).
E. Ligand Truncation
Boundary experiments were carried out to determine the minimal sequence necessary for high affinity binding ofthe RNA ligands to IFN-gamma using end-labeled RNA. Prior to end-labeling, RNA transcribed with T7 RNA polymerase was gel purified by UV shadowing. The 5'-end of20 pmoles ofeach RNA was dephosphorylated in a reaction mixture containing 20 mM Tris-Cl (pH 8.0), 10 mM MgCl2 and 0.1 U/μl shrimp alkaline phosphatase (SAP), (United States Biochemical, Cleveland OH) by incubating for
30 min at 37°C. Alkaline phosphatase activity was destroyed by incubating for 30 min at
70°C. RNA was subsequently 5'-end labeled in a reaction mixture containing 50 mM Tris-Cl (pH 7.5), 10 mM MgCl2, 5 mM DTT, 0.1 mM EDTA, 0.1 mM spermidine, 0.75 m M g -32P-ATP and 1 U/μl T4 polynucleotide kinase (New England Biolabs, Beverly MA) by incubating for 30 min at 37°C.
3'-end-labeling of20 pmoles ofeach RNA was performed in a reaction mixture containing 50 mM Tris-Cl (pH 7.8), 10 mM MgCl2' 10 mM b -mercaptoethanol, 1 mM ATP, 0.9 m M (5'-32P)pCρ and 1 U/μl T4 RNA ligase (New England Biolabs, Beverly MA) by incubating for 18 hrs at 4°C. 5'- and 3'- end-labeled RNAs were gel band purified on a 12%, 8M urea, polyacrylamide gel. After partial alkaline hydrolysis ofthe end-labeled RNA by addition ofNa2CO2 to a final concentration of50 mM and incubation in a boiling water bath for 3 min, radiolabeled RNA ligands were incubated with
IFN-gamma at three different protein concentrations, 1) 5-fold below the approximate Kd, 2) at the approximate Kd, and 3) 5-fold above the approximate Kd. Protein-bound RNA was separated by nitrocellulose partitioning. RNA truncates were analyzed on a high-resolution denaturing 12% polyacrylamide gel. To orient the sequences, a ladder of radioactively labeled ligands terminating with G-residues was generated by RNase T1 digestion ofend-labeled RNA. The T1 digest was carried out in a reaction mixture containing 7 M urea, 20 mM sodium citrate (pH 5.0), 1 mM EDTA and 5 units RNase T1 (Boehringer Mannheim, Indianapolis IN) by incubating for 5 min at 50°C.
Complementary single-stranded DNA oligonucleotides containing the sequence ofthe
T7 promoter (5'-TAATACGACTCACTATAG-3'; fragment of SEQ ID NO:2) and the sequence ofthe truncated ligand were annealed to form a double-stranded template for transcription ofeach truncated ligand. F. Receptor Binding Competitions
Human lung carcinoma cells (A549; ATCC) were plated in 24-well plates at a density of5 X 10' cells/well in RPMI 1640 plus 10% fetal bovine serum (FBS) and incubated overnight or until confluent. The cells were washed 3 times with PBS. Growth media was replaced with 200 μl RPMI 1640 plus 0.2% human serum albumin/0.02% sodium azide/20 mM Hepes, pH 7.4 together with increasing amounts (20 pg/ml-100 ng/ml) of 125I-IFN-gamma (New England Nuclear) with or without an excess (200 fold) of
unlabeled IFN-gamma. Incubations were carried out at 4°C with shaking for 2 hrs. The cells were washed 2 times with cold PBS to remove free IFN and detached with 0.5% SDS. Cell-associated 125I-IFN-gamma was determined by measuring the radioactivity of the detached cells in a gamma counter. The data was corrected for nonspecific binding and the affinity of 125I-IFN-gamma was determined by Scatchard analysis ofthe binding data. Scatchard analysis suggests that there are high-affinity binding sites (Kd = 20pM) and low-affinity binding sites (Kd = 0.5nM). For competition with oligonucleotide, the cells were incubated for 2 hr at 4°C as above with 30 pM 125I-IFN-gamma and increasing concentrations (1.01-500 nM) ofcompetitor oligonucleotide. Cell-associated
125I-IFN-gamma was determined as above.
EXAMPLE 2. 2-NH2 AND 2-F-MODIFIED RNA LIGANDS TO IFN-GAMMA
A. SELEX
Three libraries ofRNAs modified at the 2' position ofpyrimidines, 1) 2'F incorporating 2'F-CTP and 2'F-UTP, 2) 2'F/NH2 incorporating 2'F-CTP and 2'NH2-UTP and 3) 2'NH2 incorporating 2'NH2-CTP and 2'NH2-UTP were used in simultaneous SELEX protocols to generate a diverse set ofhigh-affinity modified RNA ligands to human IFN-gamma. Each ofthese libraries contained between 1013-1014 molecules with a variable region of40 nucleotides The template and primers used for the SELEX and the conditions ofthe SELEX, as described in Example 1, are summarized in Tables 1 and 2, respectively.
B. RNA Sequences and Dissociation Constants
The random modified RNA pools bound human IFN-gamma with approximate Kds ofgreater than 0.7 μM. After 17 rounds ofSELEX, the approximate Kds ofthe evolving pools had improved to, 1) 70 nM for the 2'F SELEX, 2) 115 nM for the 2'F/NH2 SELEX, and 3) 20 nM for the 2'NH2 SELEX. For mouse IFN-gamma, the approximate Kds ofthe RNA pools after 17 rounds ofSELEX were 1) 410 nM for the 2'F SELEX, 2) 175 nM for the 2'F/NH2 SELEX, and 3) 85 nM for the 2'NH2 SELEX. These Kds did not shift further in subsequent rounds.
In order to determine to what extent the evolvingpool was still random, PCR product from the final round of SELEX was sequenced as detailed above and found to be non-random. RNA from the 17th round was reverse transcribed, amplified and cloned. The sequences of 32 of the 2'F, 40 of the 2'NH2, and 11 ofthe 2'F/NH2 individual clones were determined (Table 3; SEQ IDNOS:7-65). The sequences were analyzed for conserved sequences and aligned by this criterion (Table 3). The 2'F sequences fell into 2 groups with 9 orphan sequences. Group 12'F RNAs werethe mostabundant, representing 18 of 32 sequences, while group 22'F RNAsrepresented 5 of 32 sequences. The 2'NH2 sequences fell into 2 groups with 25 of 402'NH2 RNAs in group 1 and 15 of 402'NH2 RNAs ingroup 2. The 2'F/NH2 sequences were ofa single group.
The Kds ofindividualRNAswithineachgroupweredeterminedbynitrocelulose filterbinding as described above. The Kds were determined using either amonophasic or biphasic least squares fit of the data.
Minimal sequence requirements forhigh-affinitybinding ofthe best clones were determined by 5' and 3' boundary experiments as described. The truncated RNAs were transcribed from double-stranded templates containingthe T7 promoter and the truncated sequence. Forthose successfultranscriptions, the Kd ofthetruncated ligandwas determined. The sequence of the truncated ligands and their Kds, both for full-length and forthetruncate (ifdetermined) are shown inTable 4 (SEQIDNOS:66-73).
C. Receptor Competition
Bothfull-length2'NH2 (2'NH2random,2'NH2-17,2'NH2-30)and2'F (2'F random, 2'F-1, and 2T-28) oligonucleotides were tested for their ability to inhibit receptor binding. This competition wastargetedprimarily to the high-affinity binding component using a concentration of 125I-IFN-gamma of 30pM. At this concentration, neither the 2 'NH2 nor the 2'F random oligos showed inhibition, while varying degrees ofinhibition were seen with the 4 clones tested. The 2'NH2ligand #30 (SEQ IDNO:72) was the best inhibitor and showed 50% inhibition at 10 nM.
EXAMPLE 3. EXPERIMENTAL PROCEDURES FOR 2'-NH2 AND
2-F-MODIFIED LIGANDS TO IL-4
This Example provides general procedures followed and incorporated in Example 4 for the evolution ofnucleic acid ligands to IL-4.
A. Oligonucleotides
2'F modified CTP and UTP were prepared according to the method ofPieken etal., 1991. 2'NH2 modified CTP and UTP were prepared according to the method ofMcGee et al., U.S. Patent Application No.08/264,029, filed June 22, 1994, which is incorporated herein by reference (see also McGee et al.1995). DNA oligonucleotides were synthesized by Operon Technologies (Alameda CA).
B. SELEX
The SELEX procedure has been described in detail in U.S. Patent No. 5,270,163 (see also Tuerk and Gold, 1990; Gold et al, 1993). Three SELEX procedures were performed to evolve high affinity ligands to IL-4. Each SELEX procedure utilized RNA pools containing pyrimidines modified at the 2' position as follows, 1) 2'F-CTP and 2'F-UTP referred to as 2'F, 2) 2'F-CTP and 2'NH2-UTP referred to as 2'F/NH2, and 3) 2'NH2-CTP and 2'NH2-UTP referred to as 2'NH2. For each SELEX, the DNA template 40N8 was designed to contain 40 random nucleotides, flanked by 5' and 3' regions offixed sequence (Table 5; SEQ ID NO:74). The fixed regions include DNA primer annealing sites for PCR and cDNA synthesis as well as the consensus T7 promoter region to allow in vitro transcription.
Single-stranded DNA primers and templates were synthesized and amplified into double-stranded transcribable templates by PCR. Preparation ofthe initial pool ofRNA molecules involved PCR amplification of 1000 pmoles ofsingle-stranded template (Table 5; SEQ ID NO:74) and 2500 pmoles ofboth the 5' (5P8; SEQ ID NO:75) and 3' (3P8; SEQ ID NO:76) primers. These were incubated in a reaction mixture containing 50 mM KCl, 10 mM Tris-Cl (pH 8.3), 3 mM MgCl2, 0.5 mM ofeach dATP, dCTP, dGTP, and dTTP. Taq DNA Polymerase (Perkin-Elmer, Foster City CA) at 0.1 U/μl was added and the reaction incubated at 97°C for 3 min to denature the template and primers. Following
the initial denaturing step, the reaction was cycled 7 times at 93°C for 30 sec, 53°C for 30 sec, and 72°C for 1 minto denature, anneal, and extend, respectively, the primers and template. To get an accurate concentration ofdouble-stranded PCR product forthe initial round ofSELEX, the PCR product was purifiedusing QIAquick-spin PCRpurification columns (QIAGEN Inc., Chatsworth CA) as specifiedbythe manufacturer.
For in vitro transcription using modified nucleotides 200 pmoles (final
concentration of 1 μM) of double-stranded DNA template was incubated in areaction mixture containing 40 mM Tris-Cl (pH 8.0), 12 mM MgCl2, 5 mM DTT, 1 mM
spermidine, 0.002% TritonX-100, 4% PEG 8000, 0.5 μM α-32P 2'OH ATP, 5 U/μl T7 RNA Polymerase (Davanloo et al., 1984), and concentrations of other nucleotides as follows, 1) for the 2'F SELEX: 1 mM ATP and GTP, 3 mM 2'F-CTP and 2'F-UTP, 2) for the 2'F/NH2 SELEX: 1 mM ATP, GTP, and 2'NH2-UTP and 3 mM 2'F-CTP, and 3) for the 2'NH2 SELEX: 1 mMATP, GTP, 2'NH2-CTP, and 2'NH2-UTP. These incubations were performed in a 37°C incubator for between 6 hrs and overnight. Typically the RNA was ptirifiedby gel purification and elution. To expeditetheprocess forrounds 11, 12, and 14-17 the RNA was purified using Bio-Spin 6 chromatography columns (Bio-Rad Laboratories, Hercules CA) according to manufacturer's specifications. To reduce background, the RNAwas pre-filteredpriorto all rounds ofSELEX exceptrounds 1,2,4, 6, 14, and 16. The pre-filtration step involvedbringing the RNA up to 200 μl inphosphate buffered saline (PBS), modifiedto contain 1 mM Mg2+ ions, (138 mMNaCl, 2.7 mM KCl,
8.1 mMNa2HPO4, 1.1 mM KH2PO4, ImM MgCl2, pH 7.4), (mPBS), andpassing this RNA solution through three filter discs (0.45 μm, nitrocellulose/ cellulose acetate, Millipore Corporation, Bedford MA) pre-wetted with mPBS.
For initial binding, 1000 pmoles of RNA were incubated with human IL-4 protein inbinding buffer, (mPBS plus 0.01%human serum albumin (HSA)), for 5-10 min at 37 °C to allowbinding to occur. Human recombinant IL-4 used in this SELEX procedure was purchased from R&D Systems, Minneapolis MN. Foreach round of SELEX the concentration ofRNA and protein was carefully chosen to provide optimum stringency. Preliminary experiments had shown that IL-4 had a tendency to aggregate at high protein concentrations. To prevent the evolution ofRNA species having an affinity forthis aggregated IL-4, beginning with round 4 ofSELEX and for all subsequent rounds of the
SELEX procedure, the binding mix was centrifuged at 16,000 X g for 3 min in an eppendorfcentrifuge before nitrocellulose filterpartitioning. IL-4/ RNA complexes were separated fromunbound RNAby nitrocellulose filterpartitioning described below.
Fornitrocellulose partitioning, the 2'F and 2'F/NH2 SELEX procedures used 0.2 μm pore sizepure nitrocellulose filters (Scleicher & Schuell, KeeneNH) forthe first two rounds of SELEX. All subsequent rounds of these two SELEX procedures and the entire 2'NH2 SELEX were performed with 0.45 μm pore size nitrocellulose/cellulose acetate mixed matrix filters (Millipore Corporation, BedfordMA). Filter discs were placed into a vacuummanifold andwetted with 5 ml ofmPBS buffer. The IL-4/RNA binding mix was aspiratedthroughthe filterdiscs whichwere immediatelywashedwith 5 ml of mPBS buffer. To further increase stringency and reduce background forrounds 8-13, and 15, this washing step was modifiedto include washing ofthe filter discs with 15 ml 0.5 M urea followed by 20 ml mPBS buffer. Bound RNA was isolated from filters by extraction in a solution of 400 μl phenol (equilibrated in Tris-Cl, pH 8.0)/ 300 μl 7 M urea (freshly prepared). The filters were bathed inthe phenol/urea solution atroom temperature for 30 min and at 95°C for 2 min. The RNA was phenol/chloroform extracted and ethanol precipitatedwith 20 μg tRNA.
The RNA was reversetranscribed into cDNA by addition of50 pmoles DNA primer, 0.4 mM each ofdNTPs, and 1 U/μl AMV reverse transcriptase (AMV RT) (Life Sciences, Inc., St. Petersburg FL) inbuffercontaining 50 mM Tris-Cl (pH 8.3), 60 mM NaCl, 6 mM Mg(OAc)2, 10 mMDTT. Thereactionwas incubatedat 37°C for30 min then48°C for 30 minthen 70°C for 10 min, to ensure the melting of secondary structure present inthe isolated RNA.
To begin anewround ofSELEX, the cDNA was PCR amplified by addition of250 pmoles ofboththe 5' (5P8; SEQ IDNO:75) and 3' (3P8; SEQ ID NO:76) primer in reaction conditions identical to those detailed above. The number ofcycles of PCR required to amplify the cDNA was carefully calculated for eachround ofSELEX so that 250 pmoles double-stranded DNA template would be used to initiate the next round of SELEX.
C. Equilibrium Dissociation Constants (Kds)
The determination of equilibrium dissociationconstants (Kds) for RNA pools was made subsequentto rounds 5, 8, 12, and 17 to monitortheprogress ofeach SELEX. The Kds of RNA pools formouse IL-4 (R&D Systems, Minneapolis MN) were also determined after round 8. Kds were determined for individual ligands after cloning and sequencing of RNA pools andtruncations (describedbelow). Nitrocellulose filter binding wasusedto determine Kds as follows: filter discs were placed into avacuummanifold and wetted with 5 ml of mPBS buffer.32P-labeled-RNAwas incubatedwith serial dilutions of IL-4 in binding buffer for 5-10 min at 37°C to allow binding to occur.
Binding mixes were centrifuged as described above to remove aggregates, aspirated throughthe filterdiscs, and then immediately washed with 5 ml mPBS buffer. The filter discs were dried and countedin aliquid scintillation counter (Beckmann Instruments, Palo Alto CA). Equilibrium dissociation constants were determinedby least square fitting of the datapoints using the Kaleidagraph™ graphics program (Synergy Software, Reading PA). Many ligands and evolved RNA pools yieldbiphasic binding curves. Biphasic binding canbe describedas the binding oftwo affinity species that are not in equilibrium. Biphasicbinding constantswere calculatedaccording to standardprocedures. Kds were determined by least square fitting of the data points using the Kaleidagraph™ graphics program.
D. Cloning and Sequencing
After the 17th round of SELEX, RNA molecules were reverse transcribed to cDNA and made double-stranded by PCR amplification with primers containing recognition sites forthe restriction endonucleases Hindlll (Table 5; 5' primer 5P8H; SEQ ID NO:77) and Bam HI (Table 5; 3' primer 3P8B; SEQ IDNO:78). Usingthese restriction sites the DNA sequences were inserted directionally into the pUC19 vector. These recombinant plasmids were transformed into Epicurian coli JM109 competentcells (Stratagene, La Jolla CA). Plasmid DNA was prepared with the PERFECTprep™ plasmid DNA kit (5 prime—>3 prime, Boulder CO). Plasmid clones were sequenced using a PCR sequencing protocol (Adams et al., 1991) using PCR sequencing primer pUC19F30 (SEQ ID NO:6).
E. Ligand Truncation
Boundary experiments were carried out to determine the minimal sequence necessary for high affinity binding ofthe RNA ligands to IL-4 using end-labeled RNA. Prior to end-labeling, RNA transcribed with T7 RNA polymerase was gel purified by UV shadowing. The 5'-end of20 pmoles ofeach RNA was dephosphorylated in a reaction mixture containing 20 mM Tris-Cl (pH 8.0), 10 mM MgCl2 and 0.1 U/μl shrimp alkaline phosphatase (SAP), (United States Biochemical, Cleveland OH) by incubating for 30 min at 37°C. Alkaline phosphatase activity was destroyed by incubating for 30 min at 70°C. RNA was subsequently 5'-end labeled in a reaction mixture containing 50 mM Tris-Cl (pH 7.5), 10 mM MgCl2, 5 mM DTT, 0.1 mM EDTA, 0.1 mM spermidine, 0.75 m M g -32P-ATP and 1 U/μl T4 polynucleotide kinase (New England Biolabs, Beverly MA) by incubating for-30 min at 37°C.
3'-end-labeling of20 pmoles ofeach RNA was performed in a reaction mixture containing 50 mM Tris-Cl (pH 7.8), 10 mM MgCl2, 10 mM b -mercaptoethanol, 1 mM ATP, 0.9 μM (5'-32P)pCp and 1 U/μl T4 RNA ligase (New England Biolabs, Beverly MA) by incubating for 18 hrs at 4°C. 5'- and 3'- end-labeled RNAs were gel band purified on a 12%, 8M urea, polyacrylamide gel. After partial alkaline hydrolysis ofthe end-labeled RNA by addition ofNa2CO3 to a final concentration of50 mM and incubation in a boiling water bath for 3 min, radiolabeled RNA ligands were incubated with IL-4 at three different protein concentrations, 1) 5-fold below the approximate Kd, 2) at the approximate Kd, and 3) 5-fold above the approximate Kd. Protein-bound RNA was separated by nitrocellulose partitioning. RNA truncates were analyzed on a high-resolution denaturing 12% polyacrylamide gel. To orient the sequences, a ladder ofradioactively labeled ligands terminating with G-residues was generated by RNase Tl digestion ofend-labeled RNA. The T1 digest was carried out in a reaction mixture containing 7 M urea, 20 mM sodium citrate (pH 5.0), 1 mM EDTA and 5 units RNase Tl (Boehringer Mannheim, Indianapolis IN) by incubating for 5 min at 50°C.
Complementary single-stranded DNA oligonucleotides containing the sequence ofthe T7 promoter (5'-TAATACGACTCACTATAG-3'; fragment ofSEQ ID NO:75) and the sequence ofthe truncated ligand were annealed to form a double-stranded template for transcription ofeach truncated ligand.
F. Receptor Competition
HumanT-cell lymphomacells (H-9; ATCC) were cultured in suspension inRPMI 1640 + 10% FCS. Cells were washed two times withPBS andresuspended (5.0 x 105 cells) in 200 μl media containing RPMI 1640+ 0.02% human serum albumin/0.2%Na azide/20 mM HEPES, pH 7.4 for 2 hr at4°C in 1.5 ml polypropylene tubes (Eppendorf, W. Germany) with various amounts of 125I-rIL-4 in the presence or absence of a 200-fold excess ofunlabeledcytokine. Following incubation, thetubeswere spun(150 x g, 5 min, 4°C) and the supernatant was aspirated. The cell pellet was resuspended in200 μl RPMI-HSA.100 μl aliquots were centrifugedthrough acushion ofan equal volume of phthalate oils (dibutyl/dioctyl, 1:1 v/v). The tube was rapidly frozen in dry ice/ethanol and the tip containing the cell pellet was cut off and placed in a vial for gamma counting. The data was corrected for nonspecific binding and the affinity of 125I-IL-4 was determined by Scatchard analysis. For competition with oligonucleotide, orneutralizing antibody (R & D Systems), the cells were incubated for 2 hr at 4° as above with 0.7 nM 125I-IL-4 and increasing concentrations (0.01-500 nM) of competitor oligonucleotide. Cell-associated 125I-IL-4 was determined as above.
EXAMPLE 4. 2 -NH2 AND 2 -F-MODIFIED RNA LIGANDS TO IL-4
A. SELEX
Three libraries of RNAs modified at the 2'position ofpyrimidines, 1) 2'F incorporating 2'F-CTP and2'F-UTP, 2) 2'F/NH2 incorporating 2T-CTP and 2"NH2-UTP and 3) 2'NH2 incorporating 2'NH2-CTP and2'NH2-UTP were used in simultaneous SELEX protocolsto generate adiverse set of high-affinity modified RNA ligandsto human IL-4. Each ofthese libraries contained between 1013- 1014 molecules with avariable region of 40 nucleotides. The template and primers used forthe SELEX andthe conditions ofthe SELEX, as described inExample 3 are summarized in Tables 5 and 6, respectively.
B. RNA Sequences and Dissociation Constants
The random modified RNA pools bound human IL-4 with approximate Kds of greater than 20 μM. After 17 rounds of SELEX, the approximate Kds of the evolving
pools had improved to, 1) 30 nM for the 2'F SELEX, and 2) 55 nM for the 2'F/NH2 SELEX. Binding curves performed on 2'NH2 RNA from an earlier round had shown an approximate Kd of 100 nM, however, difficulties with background reduction in this SELEX led to an apparent Kd after round 17 of 1 μM. It was felt that despite this "masking" due to background, the high affinity unique sequence 2'NH2 RNAs were still in the pool after round 17. These Kds did not shift further in subsequent rounds. The RNA pools after 8 rounds ofSELEX did not bind mouse IL-4, while there was a significant improvement in binding after 8 rounds for the human protein (data not shown).
In order to determine to what extent the evolving pool was still random, PCR product from the final round ofSELEX was sequenced as detailed above and found to be non-random. RNA from the 17th round was reverse transcribed, amplified, and cloned. The sequences of41 ofthe 2'F, 57 ofthe 2'NH2, and 30 ofthe 2'F/NH2 individual clones were determined (Table 7; SEQ ID NOS:79-177). The sequences were analyzed for conserved sequences and aligned by this criterion (Table 7). The 2'F sequences fell into a single group representing 29 of41 sequences. The remaining 12 clones were categorized as orphans due to their lack ofsequence homology with the primary group or to each other. The 2'NH2 sequences fell into 2 distinct groups ofsequences. Group 1 which represented 21 of 57 sequences were shown to bind to IL-4. The other group, representing 35 of57 sequences were shown to bind to nitrocellulose filters. The presence ofsuch a large number ofnitrocellulose filter binding RNAs was not a surprise as these sequences were cloned from a pool with high background binding. These nitrocellulose binding RNAs are identified by the presence ofa direct repeat ofthe sequence GGAGG. A single orphan 2'NH2 sequence was also found. The 2'F/NH2 sequences were more heterogeneous with sequences falling into 3 groups. RNAs in group 1 and 2 bound to IL-4, while the 3rd group bound to nitrocellulose filters. The clones in the nitrocellulose filter binding group also contained a single or repeat ofthe sequence GGAGG. It should be noted that this sequence is also found in the 3'-fixed region (underlined in Table 7).
The Kds ofindividual RNAs within each group were determined by nitrocelulose filter binding as described in Example 3 above. The Kds were determined using a monophasic least squares fit ofthe data.
Minimal sequence requirements forhigh-affinity binding of the best clones were determinedby 5' and3' boundary experiments as described inExample 3. The truncated RNAs weretranscribed from double-strandedtemplates containingthe T7 promoter and the truncated sequence. Forthose successful transcriptions, the Kd of the truncated ligand was determined. The sequence of the truncated ligands andtheirKds, both for full-length and forthetruncate (ifdetermined) are shown in Table 8 (SEQ IDNOS:178-185).
C. Receptor Competition
Full-length 2'NH2 (2'NH2 random, 2'NH2-29), 2'F (2'F random, 2'F-9) and 2'F/NH, (2'F/NH2 random, 2'F/NH2-9 and 2'F/NH2-28) oligonucleotides were tested for their ability to inhibit receptor binding. Neither the 2'NΗ2, 2'F, or 2'F/NH2 random oligos showed inhibition, while varying degrees of inhibition was seen with the clones tested. At an IL-4 concentrationof0.7 nMthe 2'F/NH2 ligand-9 was the best competitor for receptor binding and showed 50% inhibition at approximately 40 nM. The competition by this
oligonucleotide was similarto that seenby aneutralizing antibodyto IL-4.
EXAMPLE 5. EXPERIMENTAL PROCEDURES FOR2-F MODIFIED
LIGANDS TO IL-10
This Example provides general procedures followed and incorporated in Example 6 forthe evolution of nucleic acid ligands to IL-10.
A. Materials
DNA sequences were synthesized byusing cyanoethyl phosphoramidite under standard solidphase chemistry.2'-F CTP and2'-F UTP were purchased from United States Biochemicals. Human IL-10 was bought from either Bachem or R&D Systems.
Neutralizing anti-human IL-10 monoclonal antibody, murine IL-10 and ELISA detection kit forhuman IL-10 were purchased from R &D Systems.
B. SELEX
Five nmoles of synthetic DNA template, that was purified on an 8%
polyacrylamide gel under denaturing conditions were amplified by four cycles of
polymerase chain reaction (PCR). The PCR products were transcribed in vitro by T7 RNA polymerase (1000 U) in 1 mL reaction consisting of2 mM each of ATP and GTP, 3 mM each of 2'-F CTP and 2'-F UTP, 40 mM Tris-HCl (pH 8.0), 12 mM MgCl2, 1 mM Spermidine, 5 mM DTT, 0.002% TritonX-100 and 4% polyethelene glycol (w/v) for 10 - 12 hr. The full-length transcription products (SEQ ID NO: 186) were purified on 8% denaturingpolyacrylamide gels, suspended in TBS buffer [100 mM Tris-HCl, (pH 7.5) 150 mM NaCl) (binding buffer), heated to 70 °C, chilled on ice, then incubated with IL-10 at 37°C for 10 min. The RNA-proteinmixture was filteredthrough apre-wet
nitrocellulose filter then washed with 5 mL of the binding buffer. Bound RNAs were elutedfromthe filter andrecoveredby ethanol precipitation. The RNA was reverse transcribedby avianmyeloblastosis virus reverse transcriptase (Life Sciences) at 48 °C for 45 min with 5'-GCCTGTTGTGAGCCTCCTGTCGAA-3' primer (Table 9; SEQ ID NO:188). The cDNA was amplified by PCR (with 5' and 3' primers (SEQ ID
NOS:187-188)) andthe resulting DNAtemplate was transcribedto obtain RNA forthe nextround ofselection. During the course ofSELEX, the concentration ofIL-10 was decreased gradually from 5 μMto 500 nMto progressively increase selective pressure. The selectionprocess was repeateduntil the affinity of the enriched RNApool for IL-10 was substantially increased. Atthatpoint, cDNAwas amplified by PCRwithprimers that introduced BamHI and Hind III restriction sites at 5' and 3' ends, respectively. PCR products were digested with BamHI and Hind III and cloned into pUC 18 thatwas digested withthe same enzymes. Individual clones were screened and sequenced by standardtechniques.
C. Determination of equilibrium dissociation constants (Kd).
Internally-labeled RNAtranscripts were prepared by including [α-32P]ATP in T7
RNA polymerase transcription reactions. Full-lengthtranscripts were purified on 8% denaturing polyacrylamide gels to ensure size homogeneity. Gel-purified RNA was diluted to a concentration of~ 5 nM in TEM buffer, heatedto 80 °C then chilled on ice to facilitate secondary structure formation. RNA concentrations were kept lower than 100 pM in binding reactions. Briefly, equal amounts of RNA were incubated with varying amounts ofIL-10 in 50 μL ofTEM buffer for 10 min at 37°C. RNA-protein mixtures
were passed through pre-wet nitrocellulose filters (0.2 μ) and the filters were immediately washed with 5 mL of binding buffer. Radioactivity retained on filters was determined by liquidscintillation counting. The quantities of RNA bound to filters inthe absence of proteinwas determined andused forbackground correction. The percentage of input RNA retained on each filterwasplotted againstthe corresponding log protein concentration. The nonlinear least square methodto obtainthe dissociation constant (Kd).
D. Sandwich ELISA
Sandwich ELISA was carried outby using commercially available ELISA kit for quantitative determination of hIL- 10 (from R&D systems) according to manufacturer's instructions. Varying amounts ofRNA 43, randompool RNA and anti-hIL-10 monoclonal antibody (from R&D Systems) were incubated with 125 pg/mL hIL-10 at roomtemperature for 10 minbefore addedto microtiterwells. EXAMPLE 6.2'-F-MODIFIED RNA LIGANDS TO IL-10
Under nitrocellulose filter binding conditions the random sequence pool that was usedto initiate the SELEX experiment did not show detectable binding to IL-10 as high as 5 μM concentration. However, after twelve rounds ofaffinity selection the enrichedpool exhibited improved affinity, and further selection beyond the 12throundhad no effect on increasing the affinity for IL- 10. Table 10 (SEQ ID NOS : 189-205) shows the sequences identified fromthe 12throundpool. Sequences are grouped into three classes based on the sequence similarity. The 5' part inthe variable 40 nucleotide region ofmost sequences in class I has sequence complementarity to the 3' part, suggesting that such sequences can fold into astemloop structure.
Individual cloneswere initially screened fortheirabilityto bindIL-10 at250 nM concentration. The results showthat20-40% ofinput individual RNAs was bound to IL-10 at 250 nM. Based on preliminary screening, sequence 43 (SEQ ID NO:189) was chosen as arepresentative ligand to carry out in section B below.
The Kd ofsequence 43 forbinding to IL-10 is 213 nM. The ligand 43, on the otherhand does not bind to other cytokines such as interferon g and IL-4, indicating the specificity ofSELEX-derived RNA sequence. Human IL-10 (hIL-10) and mouse IL-10
(mIL-10) have high degree ofsequence homology at the cDNA and amino acid level (73% amino acid homology) and hIL-10 has been shown to active on mouse cells. However, ligand 43 does not bind to mIL-10 with high affinity. B. RNA in IL-10 ELISA
An anti-ILlO monoclonal antibody that neutralizes the receptor binding is commercially available. The R&D systems' Quantikine Immunoassay kit is based on 96 well microtiter plates coated with the neutralizing antibody to capture hIL-10. The ELISA was used to investigate whether RNA binds at or near the neutralizing antibody binding site on IL-10. RNA 43, similar to the random pool RNA (used as a control) did not show any inhibition ofIL-10 binding to anti-IL-10 antibody on the plate (data not shown). These data suggest that the evolved RNA ligand does not bind to the site at or near that recognized by the neutralizing antibody. The soluble anti-ILlO that was used in the assay as a control behaved as expected, competing for binding with the same antibody on the solid phase.
EXAMPLE 7. EXPERIMENTAL PROCEDURES FOR LIGANDS TO hTNFα
This Example provides general procedures followed and incorporated in Examples
8-11 for the evolution ofnucleic acid ligands to hTNFα.
A. Materials
Recombinant human TNFα (hTNFα ) was purchased from Genzyme (Cambridge,
MA) or R&D Systems (Minneapolis, MN), recombinant murine TNFα (mTNFα), recombinant human TNFβ (hTNFβ), and soluble human TNF receptor 2 (sTNF-R2) were purchased from R&D Systems. Acetylated, and nuclease free bovine serum albumin
(BSA), ligase and restriction enzymes were from new England Biolabs (Beverly, MA).
AMV resverse transcriptased were from Life Sciences (St. Petersburg, FL). RNasin ribonuclease inhibitor, and Taq DNA polymerase was from Promega (Madison, WI).
Ultrapure nucleotide triphosphates were from Pharmacia (Piscataway, NJ). 125-I-TNFα, α-32P-ATP, and γ-32P-ATP were from DuPont NEN Research Products (Boston, MA).
U937 cells were from ATCC (catalog number CRL1593). Oligonucleotides were obtained
from Operon, Inc. (Alameda, CA). Nitrocellulose/cellulose acetate mixed matrix (HA), 0.45 μmfilters were from Millipore (Bedford, MA). Chemicals were at least reagent grade andpurchased from commercial sources. B) SELEX
The SELEX procedure has been described in the SELEX Patent Application (see also Tuerk and Gold, 1990; Gold etal, 1993). The starting RNA contained 30 random nucleotides, flankedby 5' and 3' constantregions forprimer anealing sites for cDNA synthesis and PCR amplification (Table 11; SEQ ID NO:206). The single stranded DNA molecules were convertedto double strandedby PCRamplification. PCR conditions were 50 mM KCl, 10 mM Tris-HCl, pH9, 0.1% Triton X-100, 3 mM MgCl2, 0.5 mM of each dATP, dCTP, dGTP, and dTTP, 0.1 units/μl Taq DNA polymerase and 1 nM each of the 5' and 3' primers. Transcriptionreactions were done with about 5μMDNA template, 5 units/μl T7 RNApolymerase, 40 mM Tris-HCl (pH8), 12 mM MgCl2, 5 mM DTT, 1 mM spermidine, 0.002%TritonX-100, 4% PEG 8000, 2-4mMeach2'OHNTP, and 0.25 μM a-32P-ATP (800 Ci/mmole). For 2'F modified transcripts, 2'F-CTP and 2'F-UTP were used instead of2'OH-CTP and 2'OH-UTP. Two different SELEX experiments were done. In the first SELEX experiment, SELEX- A, the protein was immobilized onto nitrocellulose filters andthe RNA ligands werepartitionedby capture to the immobilized protein.
Briefly, hTNFα was spotted on a nitrocellulose filter (Millipore, HA 0.45 μm) and following 5 min air drying over filter paper, the nitrocellulose filter was incubated in a 24- well microtiter plate with 1-2X10-6 M radiolabeled RNA for 30 min at room
temperature in 500 μl binding buffer (BB=10 mM Tris-HCl, pH 7.5, 150 mMNaCl, 1 mM EDTA, 0.02% acetylated BSA, 0.02% ficol, and 0.02% PVP). The filterwas then washed three times for 10 minutes each in 1.5 ml BB without BSA. Binding and washing was done underrigorous agitation. The RNA boundto the immobilized protein was recovered by phenol/urea extraction and was then reverse transcribed into cDNA by AMV reverse transcriptase at 48°C for 60 min in 50 mM Tris-HCl pH8.3, 60 mM NaCl, 6 mM
Mg(OAc)2, 10 mM DTT, 50 pmol DNA primer-1 (Table 11; SEQ ID NOS:206-208), 0.4 mM each of dATP, dCTP, dGTP, and dTTP, and 1 unit/μl AMV RT. The cDNA was then PCR amplified and used to initiate the next SELEX cycle as described above. In the
second SELEX experiment, SELEX-B, the binding buffer was Dulbecco's
Phosphate-Buffered Saline (DPBS) with calcium and magnesium (Life Technologies, Gaithersburg, MD, Cat. No 21300-025) and the protein-RNA complexes were partitioned by filtering through nitrocellulose/cellulose acetated mixed matiix, 0.45 μm pore size filter disks (Millipore, Co., Bedford, MA). Nitrocellulose filterbound RNA was recovered by phenol/ureaextraction. The partitioned RNA was thenreverse transcribed and PCR amplified as above and used to initiate the next SELEX cycle.
C. Determination of Equilibrium Dissociation Constants
To partitiontheprotein-RNA complexes, the binding reactions were filtered throughnitrocellulose/cellulose acetated mixedmatrix, 0.45 μmpore size filter disks (Millipore, Co., Bedford, MA). For filtration, the filters were placed onto a vacuum manifold and wetted by aspirating 5 ml of DPBS. The binding reactions were aspirated throughtthe filters and following a 5 ml wash, the filters were counted in a scintilation counter (Beckmann). Nitrocellulose partitionsing was used for SELEX and for
determining the equilibrium dissociation constants of RNA ligands to TNFα. RNA ligands to TNFα bind monophasically.
To obtainthe equilibrium dissociation constants ofRNA ligands to TNFα the binding reaction: Ξ
is converted into an equation forthe fraction ofRNA bound at equilibrium: q=(f/2RT)(PT+RT+KD-((PT+RT+KD)2-4PTRT)1/2) q=fraction of RNA bound
PT=total protein concentration
Rτ=total RNA concentration
f=retention efficiency ofRNA-protein complexes
The average retention efficiency forRNA-TNFα complexes onnitrocellulose filters is 0.1-0.2.
The KDs were determinedby least square fitting of the data points using the software Kaleidagraph (Synergy Software, Reading, PA).
D. Cloning and Sequencing
RT-PCRamplified cDNA fromthe lastround of SELEX was cloned between
BamHI and Hindlll restriction sites of pUC18 plasmid (Vieira et al., 1982, Gene 19: 259-268) in MC1061 E. coli (Casadaban et al., 1980, JMol Biol 138: 179-207).
Sequencing was done using PCR products as templates with a commerically available kit (Promega, Madison WI).
E. Receptor Binding Competition Assay
Areceptorbinding competitionassay was usedto determinethe bioactivity of the RNA ligands.125I labelled hTNFα at 0.1 nM was incubated in 50 μl ofbinding medium (PBS with 0.5 mM Mg++ , 0.2% BSA, 0.02% sodium azide, 1U/μl RNasin) for 15 min at 4°C with serially diluted competitors at 10-4 to 10-11 M, and 1x104/μl U937 cells.
Duplicate aliquots were subsequently removed, centrifuged through 2:1
dibutyl-phthalate:dinonyl-phthalatemixtureto separate free andbound 125I labelled hTNFα, and the radioactivity inthe pelletwas measured on a gammacounter.
Nonspecific binding was determined by inclusion of a 200-fold molar excess of unlabeled TNF.
The inhibition constants (Ki) of the RNA ligands were determined by anonlinear regression analysis of the data using standard techniques. To obtain Ki values the concentration of TNF receptorwas assumed to be 3.4x10-11 M and the KD ofthe
TNFα-TNFR interaction of 0.1 nM.
F. Boundary determination
For 3' boundary determination, the 6ARNA ligand was 5' end labeledwith γ -32P-ATP using T4 polynucleotide kinase.5' boundaries were established using 3' end labeled ligand with α-32P-pCp and T4 RNA ligase. After partial alkaline hydrolysis, the radiolabeled RNA ligand was incubated withhTNFα at 5, 25, and 125 nM, and the protein bound RNA was isolated by nitrocellulose partitioning. The RNA truncates were analyzed on ahighresolution denaturingpolyacrylamide gel. An alkaline hydrolysis ladder and a ladder of radioactively labeled ligands terminated with G-residues, generated by partial RNase T1 digestion, were used as markers.
EXAMPLE 8. RNA LIGANDS TO hTNFα
A. pre-SELEX characterization
Nitrocellulose filterbinding couldnot detect any interaction of hTNFα with randomRNA even athighprotein concentrations. The binding curves were completely flat evenupto 10μM hTNFα andRNA upto lμM andthe estimated dissociation constant (KD) is greaterthan 10-3 M. No buffer conditions were found that improved the interaction of hTNFα and random RNA.
To determine whetherhTNFα was binding any RNA at all we used amore sensitivetechnique similarto northwesternprobing (Bowenetal, 1980). Thistechnique was used invarious studies ofproteinnucleic acid interaction and aided inthe cloning of various DNA binding proteins (Singh et al., 1988). This experiment showed clearly that some random RNA can bind to hTNFα. RNAbinding occurred only when the filterwas previously spotted with hTNFα and andthendried, butnot ifthe filterwas spotted with hTNFα andthenplaced wet inthe incubation chamber. The RNAwas binding only onthe filters carrying hTNFα but not on filters carrying BSA possibly because, either not enough BSA was immobilized on the filter orthe BSA present in the incubation mix was competing for available BSA-specific RNA ligands.
B. SELEX.
Two independent SELEX experiments (A and B) were initiated with pools of randomized RNA containing about 1014 unique molecules. The starting RNA and the PCRprimer sequences are shown in Table 11.
Inthe A-SELEX, the proteinwas immobilized on anitrocellulose filterby drying.
The protein containing filterwas incubated inBB (see Example 7) with labeled RNA, then washed, autoradiographed andthebound RNA was recoveredby phenol-ureaextraction. Forthe firstround ofA-SELEX about 1,000 pmoles of hTNFα monomerwas used and the RNA concentration was at 2x10-6M. For the subsequent 14 rounds, two different filters containing about 500 and 100pmoles of hTNFα monomerwere incubated inthe same chamber containing amplifiedRNA fromtheprevious round at about 2x10-6M. Only the RNA fromthe high protein filter was carried to the nextround. A steady increase inthe signal to noise ratio was observed and atround 15 the signal retained onthe 500- and 100-pmole protein filters was 170- and 35-fold above background respectively. For comparison, inthe firstroundthe signal was only about 3-fold above background. RNA fromround 15 had ahigher affinity forhTNFα with an estimated Kd of 5x10 -5 M, representing apossible 100 fold improvement overtherandomRNA. To increasethe stringency of the selection, we carried 8 more rounds using filters with about 10 and 1 pmole of hTNFα. For all these subsequent rounds, except for round 20, the RNA from the 1 pmole hTNFα filters was carried to the next round. Because of high background, at round20 we usedthe RNA from the 10 pmoles hTNFα filterofround 19. The signal to noise ratio forthese subsequent rounds became worse at eachroundbut nevertheless the affinity of the evolved RNA continued to improve with estimated final Kd of 7x10-7 M, which represents two additional orders of magnitude improvement. Inthe final round, we could detect signal with 10-fold shorterexposure time was detected, and with 100 - fold less hTNFα onthe filter.
In parallel with the stringent phase of A-SELEX, RNA from round 15 of the A-SELEX was evolved using B-SELEX conditions (see below) for 6 more rounds. We designated this as C-SELEX. The affinity of the evolved population at the end of
C-SELEX was similarto the round 23 population ofA-SELEX with approximate
Kd=4x10-7M.
The evolved RNA from round 23 had not only improved affinity for hTNFα but it was also specific (Table 13). Binding could be detected only with hTNFα.
In the B- SELEX experiment, binding reactions were set in 25-50μl and after 10 min incubation at 37°C it was filtered through a 0.45 μm HA nitrocellulose filter. For the first round ofthe B-SELEX, the RNA and protein were at about 4x10-5 M each. Under these conditions only 0.1% ofthe input RNA was retained on the filter. This was not surprising since the hTNFα-random-RNA interaction is very weak with a Kd too high to measure and probably in the 10-3M range. Subsequent rounds were set similar to the first round. By round 8, the background binding ofthe RNA to the nitrocellulose filters was very high.
C. RNA sequences and Affinitites
RT-PCR amplified cDNA from round 23 ofA-SELEX and round 6 ofC-SELEX were cloned and sequenced as described in Example 7. 37 clones were sequenced from A-SELEX and 36 cloned from C-SELEX. From the total of73 sequences, 48 were unique (Table 12; SEQ ID NOS:209-255). A unique sequence is defined as one that differs from all others by three or more nucleotides. Ofthe 47 unique clones, 18 clones could bind to hTNFα with Kd better than 1 μM (Table 12). The best ligand, 25A, (SEQ ID NO:233) binds with affinity dissociation constant at about 40 nM. Ifit is assumed that the random RNA binds with a dissociation constant ofgreater than 10-3 M, then the affinity of25A is at least four to five orders ofmagnitude better than the starting pool.
Using sequence alignment and conserved predicted secondary structure, 17 out of 18 clones that bind hTNFα could be assigned into two classes.
The members ofthe class II can be folded in stem-loop structures with internal bulges and asymmetric loops. Linear sequence alignment did not reveal any significant conserved sequences.
D. Specificity ofRNA Ligands to TNF
We tested the specificity ofthe evolved pool ofround 23 ofA-SELEX against human TNFα, human TNFβ and murine TNFα. The evolved pool is highly specific for human TNFα and specificity ratios are shown in Table 13.
EXAMPLE 9. Inhibition of hTNFα Binding to Cell Surface receptors
To testthe ability of the TNFα ligands to competitively inhibit the binding of hTNFα to its cell surface receptor, the U937 cells were usedto screen several hTNFα ligands. The observedKis are listed in Table 14. The data showthat several ligands can competitively inhibitbinding ofhTNFα to its cell surface receptors while random RNA cannot. Ligand 25Ahasthehighestpotency with aKi of21 nM. This Ki value is only 6 foldworse thanthe Ki observed withthe sTNF-R2 underthe same experimental conditions.
EXAMPLE 10. Effect of 2'F Pyrimidine Modification on the Binding and Inhibitory Activities of the hTNFα Ligands
Transcripts containing 2'F modifiedpyrimidines are resistantto RNase degradation. To obtain ligands with improved stability we tested the effect of 2'F pyrimidine modificationonthe binding and inhibitory activity ofseveral hTNFα ligands. The results summarized in Table 15 showthat some of the ligands retained binding activity when are modifiedwith 2'F pyrimidines butin general the modified ligands bind worse thanthe unmodified counterparts. Class II ligands are in general moretolerant of the 2'F pyrimidine modification. Most of the ligands thatretained binding afterthe 2'F pyrimidine modification lose their inhibitory activity. Only the 2'F pyrimidine modification of the most abundant ligand, 6A, didnot affect its binding and inhibitory activities.
EXAMPLE 11: Experimental Procedures for DNA Ligands to RANTES
This example provides general procedures followed and incorporated in Example
12 forthe evolution ofnucleic acid ligands to RANTES.
A. Materials
Recombinant human RANTES was purchased from Genzyme (Cambridge, MA). Taq DNA polymerase was Perkin Elmer (Norwalk, CT). T4 polynucleotide kinase was purchased from New England Biolabs (Beverly, MA). Ultrapure nucleotide triphosphates
were purchased form Pharmacia (Piscataway, NJ). Affinity purified streptavidin (Cat. No 21122) was from Pierce (Rockford, IL). Oligonucleotides were obtained from Operon, Inc. (Alameda, CA). Nitrocellulose/cellulose acetate mixed matrix (HA), 0.45 μm filters were purchased form Millipore (Bedford, MA). Chemicals were at least reagent grade and purchased from commercial sources.
B. SELEX
The SELEX procedure has been described in detail in the SELEX Patent
Applications. The DNA template contained 40 random nucleotides, flanked by 5' and 3' constant regions for primer anealing sites for PCR (Table 16; SEQ ID NOS:256-258). Primer 3G7 (SEQ ID NO:258) has 4 biotin residues in its 5' end to aid in the purification ofsingle stranded DNA (ssDNA). For the first round, 105 pmoles ofsynthetic 40N7 ssDNA were 5' end labelled using T4 polynucleotide kinase in a 25 μl reaction containing 70 mM Tris-HCl pH 7.6, 10 mM MgCl2, 5 mM DTT, 39.5 pmoles of g -32P-ATP (3000 Ci/mmol), and 16 units kinase, for 1 h at 37°C. The kinased DNA was then purified on an 8% polyacrilamide, 7M urea, denaturing gel and then mixed with gel purified unlabeled 40N7 to achieve about 5,000 cpm/pmol specific activity. To prepare binding reactions, the DNA molecules were incubated with recombinant RANTES in Hanks' Balanced Salt Solution (HBSS) without calcium and magnesium (Life Technologies, Gaithersburg, MD, Cat. No 14175) containing 0.01% human serum albumin. Two SELEX experiments were performed, one with normal salt concentration and the other with 300 mM NaCl. The high salt concentration was achieved by adding additional NaCl to the HBSS. Following incubation at room temperature for 30 minutes the protein-DNA complexes were partitioned from unbound DNA by filtering through HA nitrocellulose 0.45 μm.
Nitrocellulose filter bound DNA was recovered by phenol/urea extraction. The partitioned
DNA was PCR amplified in 50 mM KCl, 10mM Tris-HCl, pH9, 0.1% Triton X-100, 3mM MgCl2, 1 mM ofeach dATP, dCTP, dGTP, and dTTP, with 0.1 units/μl Taq DNA polymerase. The 3G7 and 5G7 primers were present at 2 μM. The 5G7 primer was 5'-end labeled before use described above. To purify ssDNA, the PCR product was ethanol precipitated and then reacted with affinity purified streptavidin at a molar ratio 1:10 DNA to streptavidin in 10 mM Tris-HCl pH 7.5, 50 mM NaCl, 1 mM EDTA, 0.05% sodimm
azide. Following 30 incubation atroomtemperature, equal volume of100% formamide tracking dye was addedandthe strands were denaturedby incubating at 85°C for 1.5 min. The denatured strands were then electophoresed in an 8% polyacrylamide, 7M urea gel andthenonshifted bandwas excised andpurified fromthe crushed gel. The purified ssDNAwas thenused forthe next SELEX cycle.
EXAMPLE 12: DNALIGANDS TO RANTES
A. SELEX
To generate DNA ligands forRANTES, two SELEX experiments were performed, one with 150 mM andthe otherwith 300 mMNaCl. The high salt was used in orderto avoid precipitation of the RANTES-DNA complexes that occurs at the lower salt concentration. The SELEXat300 mM saltwasprematurely terminatedbecause ofhigh background. The SELEX conditions and results for each round of the 150 mM salt SELEX are summarized inTable 17. The startingpool contained 1.8xl015 (2,940 pmoles) of DNA for the 150 mM salt SELEX. The starting KD values of the random DNA were 3x10-6M. After 19 rounds of SELEX the evolved pools bound with a KD of 20 nM. This represents about 150 fold improvement.
References
Adams, D. O., and T. A. Hamilton (1984) Annu Rev. Immunol.2:283-318.
Adams, S. M., and R. Blakesley (1991) Focus 13 (2):56-58.
Aggarwal, B. B., W. J. Kohr, P. E. Hass, B. Moffat, S. A. Spencer, W. J. Henzel, T. S.
Bringman, G. E. Nedwin, D. V. Goeddel, R. N. Harkins (1985) J. Biol. Chem. 260:2345-2354.
Alam, R., S. Stafford, P. Forsythe, R. Harrison, D. Faubic, M. A. Lett-Brown, J. A. Grant (1993) J. Immunol.150:3442-3447.
Allen, P. M., and E. R. Unanue (1987) Adv. Exp. Med. Biol.225:147-154.
Anderson, P., Y. K. Yip, J. Vilcek (1982) J. Biol. Chem.257:11301-11304.
Bancroft, G. J., R. D. Schreiber, G. C. Bosma, M. J. Bosma, E. R. Unanue (1987) J.
Immunol.139:1104-1107.
Bancroft, G. J., K. C. F. Sheehan, R. D. Schreiber, E. R. Unanue (1989) J. Immunol.
143:127-130.
Bazan, J. F. (1990) Immunol. Today 11:350-354.
Belosevic, M., D. S. Finbloom, P. H. VanderMeide, M. V. Slayter, C. A. Nacy (1989) J.
Immunol.143:266-274.
Beutler, B., and A. Cerami (1989) Ann. Rev. Immunol.7:625-655.
Beutler, B., I. W. Milsark, A. C. Cerami (1985) Science 229:869-871.
Bischoff, S. C., M. Krieger, T. Brunner, A. Rot, V. V. Tscharner, M. Baggiolini, C. A.
Dahinden(1993) Eur. J. Immunol.23:761-767.
Bowen, B., J. Steinberg, U. K. Laemmli, H. Weintraub (1980) Nucl. Acids Res.8:1-20.
Brennan, F. M., D. Chantry, A. Jackson, R. Maini, M. Feldmann (1989) Lancet
2:244-247.
Briscoe, D. M., R. S. Cotran, J. S. Pober (1992) J. Immunol.147:2954-2960.
Brockhaus, M., H. J. Schoenfeld, E. J. Schlaege, W. Hunziker, W. Lesslauer (1990) Proc. Natl. Acad. Sci. USA 87:3127-3131.
Brown, M. A., J. H. Pierce, C. J. Watson, J. Falco, J. N. Ihle, W. E. Paul (1987) Cell
50:809-818.
Buchmeier, N. A., and R. D. Schreiber (1985)Proc. Natl. Acad. Sci. USA 82:7404-7408.
Casadaban, M., and S. Cohen (1980) J. Mol. Biol.138:179-207.
Casey, L. C, R. A. Balk, R. C. Bone (1993) Ann. Intern. Med.119:771-777.
Clerici, M., and G. M. Shearer (1993) Immunol. Today 14:107-111.
Coffman, R. L., J. OΗara, M. W. Bond, J. Carry, A. Zlotnik, W. E. Paul (1986) J.
Immunol.136:4538-4541.
Crawford, R. M., D. S. Finbloom, J. OΗara, W. E. Paul, M. S. Meltzer (1987) J.
Immunol.139:135-141.
Davanloo, P., A. H. Rosenberg, J. J. Dunn, F. W. Studier (1984)Proc. Natl. Acad. Sci.
USA 81:2035-2039.
Del Prete, G., E. Maggi, P. Parronchi, I. Chraetien, A. Tiri, D. Macchia, M. Ricci, J.
Banchereau, J. DeVries, S. Romagnani (1988)J Immunol.140:4193-4198.
Derynck, R., D. W. Leung, P. W. Gray, D. V. Goedell (1982) Nucl. Acids Res.
10:3605-3615.
Dinarello, C. A., andJ. G. Cannon (1993) Ann. Intern. Med.119:853-854.
Ding, A. H., C. F. Nathan, D. J. Stuehr (1988)J. Immunol.141:2407-2412.
Duh, E. J., W. J. Maury, T. M. Folks, A. S. Fauci, A. B. Rabson (1989) Proc. Natl. Acad.
Sci. USA 86:5974-5978.
Eck, M. J., B. Beutler, G. Kuo, J. P. Merryweather, S. R. Sprang (1988) J. Biol. Chem.
263:12816-12819.
Erbe, J. E., Collins, L. Shen, R. F. Graziano, M. W. Fanger(1990) Mol. Immunol.
27:57-67.
Feldmann, M. (1989) Immunology Supplement 2:66-71.
Ferrara, J. L. M., and H. J. Deeg (1991) N. Engl. J. Med. 324:667-674.
Fiedler, V. B., I. Loof, E. Sander, V. Voehringer, C. Galanos, M. A. Fournel (1992) J. Lab. Clin. Med.120:574-588.
Finkelman, F. D., I. M. Katona, J. F. Urbon, Jr., C. M. Snapper, J. Ottara, W. E. Paul (1986) Proc. Natl. Acad. Sci. USA 83:9765-9778.
Finkelman, F. D., J. Holmes, I. M. Katona, J. F. Urban, Jr., M. P. Beckman, L. S. Park, K.
Schooley, R. L. Cofftnan, T. R. Mosmann, W. E. Paul (1990) Annu Rev. Immunol. 8:303-333.
Finkelman, F. D., I. M. Katona, J. F. Urban, Jr., W. E. Paul (1989) In Ciba Foundation Symposium. IgE, Mast Cells and the Allergic Response. Wiley. Chichester, ρρ.3-22.
Folks, T. M., K. A. Clouse, J. Justment, A. Rabson, E. Duh, J. H. Kehrl, A. S. Fauci
(1989) Proc. Natl. Acad. Sci. USA 86: 2365-2368.
Galizzi, J. P., C. E. Zuber, N. Harada, D. M. Gorman, O. Djossou, R. Kastelein, J.
Banchereau, M. Howard, A. Miyajima (1990) Int. Immunol.2:669-675.
Gao, J. L., D. B. Kuhns, H. L. Tiffany, D. McDermott, X. Li, U. Francke (1993) J. Exp.
Med.177:1421-1427.
Gazzinelli, R. T., I. P. Oswald, S. L. James, A. Sher (1992) J. Immunol.148:1792-1796.
Gleichmann, E., E. H. VanElven, J. P. W. VanderVeen (1982) Eur. J. Immunol.
12:152-159.
Gold, L., C. Tuerk, P. Allen, J. Binkley, D. Brown, L. Green, S. MacDougal, D.
Schneider, D. Tasset, S. Eddy (1993) In the RNA World. Gesteland, R. F., and J. F. Atkins (eds.). Cold Spring Harbor Laboratory Press.
Grau, G. E., L. F. Fajardo, P. F. Piguet, B. Allet, P. H. Lambert, P. Vassalli (1987)
Science 237:1210-1212.
Green, S. J., R. M. Crawford, J. T. Hockmeyer, M. S. Meltzer, C. A. Nacy (1990) J.
Immunol.145:4290-4297.
Greenlund, A. C. (1992) J. Interferon Res. (Abstr.) 12:S103.
Gu, D., L. Wogensen, N. A. Calcutt, C. Xia, S. Zhu, J. P. Merlie, H. S. Fox (1995) J. Exp.
Med.181:547-557.
Haak-Frendscho, M., S. A. Marsters, J. Mordenti, S. Brady,N. A. Gillett, S. A. Chen, A. Ashkenazi(1994) J. Imunol.152:1347-1353.
Harada, N., B. E. Castle, D. M. Gorman,N. Itoh, J. Schreurs, R. L. Barrett (1990) Proc.
Natl. Acad. Sci. USA 87:857-861.
Heinzel, F. P. M. D. Sadick, S. S. Mutha, R. M. Locksley (1991) Proc. Natl. Acad. Sci.
USA 887011-7015.
Heinzel, F. P., M. D. Sadick, B. J. Holaday, R. L. Cof&nan, R. M. Locksley (1989)J. Exp.
Med.169:59-72.
Hershey, G. K. K., D. W. McCourt, R. D. Schreiber (1990)J. Biol. Chem.
265:17868-17875.
Higuchi, C. M., J. A. Thompson, C. G. Lindgren, S. GiUis, M. B. Widmer, D. E. Kern, A.
Fefer (1989) CancerResearch 49:6487-6492.
Hinshaw, L. B., T. E. Emerson, Jr., F. B. Taylor, Jr., A. C. Chang, M. Duerr, G. T. Peer, D. J. Flournoy, G. L. White, S. D. Kosanke, C. K. Murray (1992) J. Trauma 33:568-573.
Hohmann, H. P., R. Remy, M. Brockhaus, A. P. G. M. Van-Loon (1989)J. Biol. Chem.
264:14927-14934.
Hoon, D. S. B., E. Okunn, M. Banez, R. F. Irie, D. L. Morton (1991) Cancer Research 51:5687-5693.
Horuk, R., C. E. Chitnis, W. C. Darbonne, T. J. Colby, A. Rybicki, T. J. Hadley, L. H.
Miller (1993) Science 261:1182-1184.
Howard, J. G., C. Hale, F. Y. Liew (1980) Nature 288:161-162.
Howard, M., J. Farrar, M. Hilfiker, B. Johnson, K. Takatsu, T. Hamaoka, W. E. Paul
(1982)J. Exp. Med.155:914-923.
Hsieh, C. -S., S. E. Macatonia, C. S. Tripp, S. F. Wolf, A. O'Garra, K. M. Murphy (1993) Science 260:547-549.
Hu-Li, J., E. M. Shevach, J. Mizuguchi, J. OΗara, T. Mosmann, W. E. Paul (1987) J. Exp.
Med.165:157-172.
Idzerda, R. L., C. J. March, B. Mosley, S. D. Lyman, T. V. Bos, S. D. Gimpel, W. S. Din, K. H. Grabstein, M. B. Widmer, L. S. Park, D. Cosman, M. P. Beckman (1990) J. Exp. Med.171:861-73.
Isakson, P. C., E. Pure, E. S. Vitetta, P. H. Krammer (1982) J. Exp. Med.155:734-748.
Jansen, J. H., G.-J. H. M. Wientjens, W. E. Fibbe, R. Willemze, H. C. Kluin-Nelemans (1989) J Exp. Med.170:577-582.
Jones, E. Y., D. I. Stuart, N. P. C. Walker (1989) Nature 338:225-228.
Jung, V., A .Rashidbaigi, C. Jones, J. Tischϋeld, T. B. Shows (1987) Proc. Natl. Acad.
Sci. USA 84:4151-4155.
Katona, I. M., J. F. Urban, Jr., S. S. Kang, W. E. Paul, F. Finkelman (1991) J. Immunol.
146:4215-4221.
Keffer, J., L. Probert, H. Cazlaris, S. Georgopoulos, E. Kaslaris, D. Kioussis, G. Kollias (1991) EMBOJ.10:4025-4031.
Khorram, O., R. N. Taylor, I. P. Ryan, T. J. Schall, D. V. Landers (1993) Am. J. Obstet.
Gynecol.169:1545-1549.
Kishimoto, T., T. Taga, S. Akira (1994) Cell 76:253-262.
Kϋhn, R., J. Lohler, D. Rennick, K. Rajewsky, W. Mϋller (1993) Cell 75:263-274.
Kϋhn, R., R. Rajewsky, W. Mϋller (1991) Science 254:707-710.
Kuna, P., S. R. Reddigari, T. J. Schall, D. Rucinski, M. Sadick, A. P. Kaplan (1993) J.
Immunol.150:1932-1943.
Kuna, P., S. R. Reddigari, T. J. Schall, D. Rucinski, M. Y. Viksman, A. P. Kaplan (1992) J. Immunol.149:636-642.
Landolfo, S., F. Cofano, M. Giovarelli, Maria Pratt, G. Cavallo, G. Forni (1985) Science 229:176-179.
Le Beau, M. M., R. S. Lemons, R. Espinosa III, R. A. Larson, N. Arai, J. D. Rowley (1989) Blood 73:647-50.
Leal, L. M. C. C., D. W. Moss, R. Kϋhn, W. Miller, F. Y. Liew (1993) Eur. J. Immunol.
23:566-569.
Lebman, D. A., Coffman, R. L. (1988) J. Exp. Med.168:853-862.
Lee, F., T. Yokota, T. Otsuka, P. Meyerson, D. Villaret, R. Coffman, T. Mossman, D.
Rennick, N. Roehm, C. Smith, A. Zlotnik, K-I. Arai (1986) Proc. Natl. Acad. Sci. USA 83:2061-2065.
Lenardo, M. J., and D. Baltimore (1989) Cell 58:227-229.
Lewis, D. B., H. D. Liggitt, E. L. Effman, S. T. Motley, S. L. Teitelbaum, K. J. Jepsen (1993) Proc. Natl. Acad. Sci. USA 90:11618-11622.
Liew, F. Y., S. Millott, J. Li, R. Lelchuk, Y. L. Chan, H. Ziltener (1989) Eur. J. Immunol.
19:1227-1232.
Loetscher, H., M. Steinmetz, W. Lesslauer (1991) Cancer Cells 3:221-226.
Loetscher, H., Y.-C. E. Pan, H.-W. Lahm, R. Gentz, M. Brockhaus, H. Tabuchi, W.
Lesslauer (1990) Cell 61:351-359.
Lowenthal, J. W., D. W. Ballard, E. Boehnlein, W. C. Green (1989) Proc. Natl. Acad. Sci.
USA 86:2331-2335.
Lowenthal, J. W., B. E. Castle, J. Christiansen, J. Schreuers, D. Rennick, N. Arai, P. Hoy, Y. Takebe, M. Howard (1988) J Immunol.140:456-464.
Mantovani, A., and E. Dejana (1989) Immunol. Today 10:370-375.
McGee, D. P. C., C. Vargeese, Y. Zhai, G. P. Kirschenheuter, A. Settle, C. R. Siedem, W.
A. Pieken (1995) Nucleosides & Nucleotides 14:1329-1339.
Mclnnes, A., and D. M. Rennick (1988) J. Exp. Med.167:598-611.
Mogi, M., M. Harada, P. Riederer, H. Narabayashi, K. Fujita, T. Nagatsu (1994) Neurosc.
Lett.165:208-210.
Mohler, K. M., D. S. Torrance, C. A. Smith, R. G. Goodwin, K. E. Stremler, V. P. Fung, H. Madani, M. B. Widmer (1993) J. Immunol.151:1548-1561.
Moore, K. W., P.Vieira, D. F. Fiorentino, M. L. Trounstine, T. A. Kahn, T. R. Mosmann (1990) Science 248:1230-1234.
Moore, K. W., A. O'Garra, R. deWaal Malefyt, P. Vieira, T. R. Mossman (1993) Ann.
Rew. Immunol.11:165-190.
Mosely, B., M. P. Beckman, C. J. March, R. L. Idzerda, S. D. Gimpel, T. VandenBos, D.
Friend, A. Alpert, D. Anderson, J. Jackson, J. M. Wignall, C. Smith, B. Gallis, J. E. Sims, D. Urdal, M. B. Widmer, D. Cosman, L. S. Park (1989) Cell 59:335-348.
Mosmann, T. R., M. W. Bond, R. L. Coffman, J. OΗara, W. E. Paul (1986) Proc. Natl.
Acad. Sci. USA 83:5654-5658.
Mosmann, T. R., and R. L. Coffm an (1989) Annu. Rev. Immunol.7:145-173.
Mulkerrin, M. G., and R. Wetzel (1989) Biochemistry 28:6556-6561.
Munro, J. M., J. S. Pober, R. S. Cotran (1989) Am. J. Path.135:121-131.
Nelson, P. J., H. T. Kim, W. C, Manning, T. J. Goralski, A. M. Krensky (1993) J.
Immunol.151:2601-2612.
Neote, K., W. Darbonne, J. Ogez, R. Horuk, T. J. Schall (1993) J. Biol. Chem.
268:12247-12249.
Noma, Y., P. Sideras, T. Naito, S. Bergstedt-Lindquist, C. Azuma, E. Severinson, T.
Tanabe, T. Kinashi, F. Matsuda, Y. Yaoita, T. Honjo (1986) Nature 319:640-46.
Oliff, A., D. Defeo-Jones, M. Boyer, D. Martinez, D. Kiefer, G.Vuocolo, A. Wolfe, S. H.
Socher(1987) Cell 50:555-563.
Oliver, K., R. J. Noelle, J. W. Uhr, P. H. Krammer, E. S. Vitetta (1985) Proc. Natl. Acad.
Aci. USA 82:2465-2467.
Osborn, L., S. Kunkel, G. J. Nabel (1989) Proc. Natl. Acad. Sci. USA 86:2336-2340.
Park, L. S., D. Friend, K. Grabstein, D. L. Urdal (1987) Proc. Natl. Acad. Sci. USA
84:1669-1673.
Parrillo, J. E. (1991) Annals Int. Med.115:491-492.
Paul, W. E. (1991) Blood 77:1859-870.
Paul, W. E., and R. A. Seder (1994) Cell 76:241-251.
Pennica, D., V. T. Lam, N. K. Mize, R. F. Weber, M. Lewis, B. M. Fendley, M. T. Lipari, D. V. Goeddel (1992) J. Biol. Chem.267:21172-21178.
Pieken, W., D. B. Olsen, F. Benseler, H. Aurup, F. Eckstein (1991) Science 253:314-317.
Piguet, P. F. (1992) TNF and Alloreactions. Involvement ofTNF in the Effector Phase of Graft-Versus-Host and host-Versus-Graft Reactions. TUMOR NECROSIS FACTORS: The Molecules and Their Emerging Role in Medicine. New York, Raven Press, Ltd.
Plaut, M., J. H. Pierce, C. J. Watson, J. Hanley-Hyde, R. P. Nordan, W. E. Paul (1989) Nature 339:64-67.
Pober, J. S., M. A. Gimbrone, Jr., L. A. Lapierre, D. L. Mendrick, W. Fiers, R. Rothlein, T. A. Springer (1986)J. Immunol.137:1893-1896.
Rabin, E., J. OΗara, W. E. Paul (1985)Proc. Natl. Acad. Sci. USA 82:2935-2939.
Rapoport, M. J., A. Jaramillo, D. Zipris, A. H. Lazarus, D. V. Serreze, E. H. Leiter, P.
Cyopick, J. S. Danska, T. L. Delovitch (1993) J. Exp. Med.178:87-99.
Rathanaswami, P., M. Hachicha, M. Sadick, T. J. Schall, S. R. McColl (1993) J. Biol Chem.268:5834-5839.
Ravetch, J. V. (1994) Cell 78:553-560.
Rinderknecht, E., B. H. O'Connor, H. Rodriguez (1984)J. Biol. Chem.259:6790-6797.
Rot, A., M. Krieger, T. Brunner, S. C. Bischoff, T. J. Schall, C. A. Dahinden (1992) J.
Exp. Med.176:1489-1495.
Salgame, P., J. S. Abrams, C. Clayberger, H. Goldstein, J. Convit, R. L. Modlin, B. R.
Bloom (1991) Science 254:279-282.
Sato, T. A., M. B. Widmer, F. D. Finkelman, H. Madani, C. A. Jacobs, K. H. Grabstein, C.
R. Maliszewski (1993)J. Immunol.150:2717-2723.
Scahill, S. J., R. Devos, J. VanDerHeyden, W. Fiers (1983) Proc. Natl. Acad. Sci. USA 80:4654-4658.
Schall, T. J. (1991) CYTOKINE3:165-183.
Schall, T. J., K. Bacon, K. J. Toy, D. V. Goeddel (1990) Nature 347:669-671.
Schall, T. J., M. Lewis, J. Koller, A. Lee, G. C. Rice, G. H. Wong, T. Granger, G. A.
Granger, R. Lentz, H. Raab (1990) Cell52:415-423.
Schall, T. J., N. J. Simpson, J. Y. Mak (1992) Eur. J. Immunol.22:1477-1481.
Schreiber, R. D., and A. Celada (1985) Lymphokines.11:87-118.
Schreiber, R. D., M. A. Farrar, G. K. Hershey, J. Fernandez-Luna (1992) Int. J.
Immunopharmac.14:413-419.
Shanafelt, A. B., A. Miyajima, T. Kitamura, R. A. Kastelein (1991) EMBOJ.
10:4105-4112.
Sher, A, and R. L. Coffman (1992) Ann. Rev. Immunol.10:385-409.
Sher, A., R. T. Gazzinelli, I. P. Oswald, M. Clerici, M. Kullberg, E. J. Pearce, J. A.
Berzofsky, T. R. Mosmann, S. L. James, H. L. Morse (1992) Immunol. Rev.
127:183-204.
Singh, H., J. H. LeBowitz, A. S. J. Baldwin, P. A. Sharp (1988) J. Biol. Chem.
262:6951-6954.
Snapper, C. M., and W. E. Paul (1987) Science 236:944-947.
Smith, R. A., and C. Baglioni (1987) J. Biol. Chem.262:6951-6954.
Spits, H., H. Yssel, Y. Takebe, N. Arai, T. Yakota, F. Lee, K.-I. Arai, J. Banchereau, J. E.
DeVries(1987) J. Immunol.139:1142-1147.
Steinman, L. (1993) Scientific American 269 (3):107-114.
Strunk, R. C., F. Sessions Cole, D. H. Perlmutter, H. R. Cohen (1985) J. Biol. Chem.
260:15280-15285.
Suzuki, Y., M. A. Orellana, R. D. Schreiber, J. S. Remington (1988) Science 240:516-518.
Sylvestre, D. L., Ravetch, J. V. (1994) Science 265:1095-1098.
Te Velde, A. A., J. P. G. Klomp, B. A. Yard, J. E. DeVries, C. G. Figdor (1988) J.
Immunol.140:1548-1554.
Tepper, R. I., P. K. Pattengale, P. Leder (1989) Cell 57:503-512.
Tepper, R. I., D. A. Levinson, B. Z. Stanger, J. Campos-Torres, A. K. Abbas, P. Leder (1990) Cell 62:457-467.
Thornhill, M. H., S. M. Wellicome, D. L. Mehiouz, J. S. S. Lanchbury, U. Kyan-Aung, D.
O. Haskard (1991) J. Immunol.146:592-598.
Traugott, U., and P. Lebon (1988) J. Neurol. Sci.84:257-264.
Trenn, G., H. Takayama, J. Hu-Li, W. E. Paul, M. V. Sitkovsky (1988) J. Immunol.
140:1101-1106.
Trinchieri, G., and B. Perussia (1985) Immunol. Today 6:131-135.
Tuerk, C., and L. Gold (1990) Science 249:505-510.
Umland, S. P., S. Razac, D. K. Nahrebne, B. W. Seymour (1992) Clin. Immunol.
Immunophathol. 63:66-73.
Ushiyama, C., T. Hirano, H. Miyajima, K. Okumura, Z. Ovary, H. Hashimoto (1995)J.
Immunol.154:2687-2696. vanLeeuwen, B. H., M. E. Martinson, G. C. Webb, I. G. Young (1989) Blood
73:1142-1148. vanRappard-vanDerVeen, F. M., A. G. Rolink, E. Gleichmann (1982) J. Exp. Med.
155:1555-1560.
Van-Riper, G., S. Sicilianl, P. A. Fischer, R. Meurer, M. S. Springer, H.Rosen (1993) J.
Exp. Med.177:851-856.
Vassalli, P. (1992)Annu. Rev. Immunol.10:411-452.
Via, C. S., and G. M. Shearer (1988) Immunol. Today 9:207-212.
Vieira, J., andJ. Messing (1982) Gene 19:259-268.
Vieira, P., R.deWaal-Malefyt, M. N. Dang, K. E. Johnson, R. Kastelein, D. F. Fiorentino, J. E. DeVries, M. G. Roncarlo, T. R. Mosmann, K. W. Moore (1991) Proc. Natl. Acad. Sci. USA 88:1172-1176.
Visvanathan, K. V., and S. Goodbourn (1989) EMBOJ.8:1129-1138.
Vitetta, E. S., J. O'H ara, C. D. Myers, J. E. Layton, P. H. Krammer, W. E. Paul (1985)J.
Exp. Med.162:1726-1731.
Walter, M. R., W. J. Cook, B. G. Zhao, R. P. Cameron, Jr., S. E. Ealick, R. L. Walter, Jr., P. Reichert, T. L. Naqab-hushan, P. P. Trotta, C. E. Bugg (1992)J. Biol. Chem. 267:20371-20376.
Wang, J. M., D. W. McVicar, J. J. Oppenheim, D. J. Kelvin (1993)J. Exp. Med.
177:699-705.
Wheelock, E. F. (1965)Science 149:310-311.
Widmer, M. B., and K. H. Grabstein (1987) Nature 326:795-798.
Wiedermann, C. J., E. Kowald, N. Reinisch, C. M. Kaehler, I. von-Luettichau, J. J.
Pattison, P. Huie, R. K. Sibley, P. J. Nelson, A. M. Krensky (1993) Current Biology3:735-739.
Williams, R. O., M. Feldmann, R. N. Maini (1992) Proc. Natl. Acad. Sci. USA 89:9784-9788.
Yokota, T., T. Otsuka, T. Mossman, J. Banchereau, T. DeFrance, D. Blanchard, J. E.
DeVries, F. Lee, K.-I. Arai (1986)Proc. Natl. Acad. Sci. USA 83:5894-5898.